diff --git a/demo-notebooks/guided-demos/0_basic_ray.ipynb b/demo-notebooks/guided-demos/0_basic_ray.ipynb new file mode 100644 index 000000000..7daf4b88b --- /dev/null +++ b/demo-notebooks/guided-demos/0_basic_ray.ipynb @@ -0,0 +1,205 @@ +{ + "cells": [ + { + "attachments": {}, + "cell_type": "markdown", + "id": "8d4a42f6", + "metadata": {}, + "source": [ + "In this first notebook, we will go through the basics of using the SDK to:\n", + " - Spin up a Ray cluster with our desired resources\n", + " - View the status and specs of our Ray cluster\n", + " - Take down the Ray cluster when finished" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b55bc3ea-4ce3-49bf-bb1f-e209de8ca47a", + "metadata": {}, + "outputs": [], + "source": [ + "# Import pieces from codeflare-sdk\n", + "from codeflare_sdk.cluster.cluster import Cluster, ClusterConfiguration\n", + "from codeflare_sdk.cluster.auth import TokenAuthentication" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "614daa0c", + "metadata": {}, + "outputs": [], + "source": [ + "# Create authentication object for oc user permissions\n", + "auth = TokenAuthentication(\n", + " token = \"XXXXX\",\n", + " server = \"XXXXX\",\n", + " skip_tls=False\n", + ")\n", + "auth.login()" + ] + }, + { + "cell_type": "markdown", + "id": "bc27f84c", + "metadata": {}, + "source": [ + "Here, we want to define our cluster by specifying the resources we require for our batch workload. Below, we define our cluster object (which generates a corresponding AppWrapper)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0f4bc870-091f-4e11-9642-cba145710159", + "metadata": {}, + "outputs": [], + "source": [ + "# Create and configure our cluster object (and appwrapper)\n", + "cluster = Cluster(ClusterConfiguration(\n", + " name='raytest',\n", + " namespace='default',\n", + " min_worker=2,\n", + " max_worker=2,\n", + " min_cpus=1,\n", + " max_cpus=1,\n", + " min_memory=4,\n", + " max_memory=4,\n", + " gpu=0,\n", + " instascale=False\n", + "))" + ] + }, + { + "cell_type": "markdown", + "id": "12eef53c", + "metadata": {}, + "source": [ + "Next, we want to bring our cluster up, so we call the `up()` function below to submit our cluster AppWrapper yaml onto the MCAD queue, and begin the process of obtaining our resource cluster." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f0884bbc-c224-4ca0-98a0-02dfa09c2200", + "metadata": {}, + "outputs": [], + "source": [ + "# Bring up the cluster\n", + "cluster.up()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "657ebdfb", + "metadata": {}, + "source": [ + "Now, we want to check on the status of our resource cluster, and wait until it is finally ready for use." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3c1b4311-2e61-44c9-8225-87c2db11363d", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.status()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "a99d5aff", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.wait_ready()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "df71c1ed", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.status()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "b3a55fe4", + "metadata": {}, + "source": [ + "Let's quickly verify that the specs of the cluster are as expected." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7fd45bc5-03c0-4ae5-9ec5-dd1c30f1a084", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.details()" + ] + }, + { + "cell_type": "markdown", + "id": "5af8cd32", + "metadata": {}, + "source": [ + "Finally, we bring our resource cluster down and release/terminate the associated resources, bringing everything back to the way it was before our cluster was brought up." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5f36db0f-31f6-4373-9503-dc3c1c4c3f57", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.down()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0d41b90e", + "metadata": {}, + "outputs": [], + "source": [ + "auth.logout()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.9.13" + }, + "vscode": { + "interpreter": { + "hash": "f9f85f796d01129d0dd105a088854619f454435301f6ffec2fea96ecbd9be4ac" + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/demo-notebooks/guided-demos/1_basic_instascale.ipynb b/demo-notebooks/guided-demos/1_basic_instascale.ipynb new file mode 100644 index 000000000..8bb86e125 --- /dev/null +++ b/demo-notebooks/guided-demos/1_basic_instascale.ipynb @@ -0,0 +1,178 @@ +{ + "cells": [ + { + "attachments": {}, + "cell_type": "markdown", + "id": "9865ee8c", + "metadata": {}, + "source": [ + "In this second notebook, we will go over the basics of using InstaScale to scale up/down necessary resources that are not currently available on your OpenShift Cluster (in cloud environments)." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b55bc3ea-4ce3-49bf-bb1f-e209de8ca47a", + "metadata": {}, + "outputs": [], + "source": [ + "# Import pieces from codeflare-sdk\n", + "from codeflare_sdk.cluster.cluster import Cluster, ClusterConfiguration\n", + "from codeflare_sdk.cluster.auth import TokenAuthentication" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "614daa0c", + "metadata": {}, + "outputs": [], + "source": [ + "# Create authentication object for oc user permissions\n", + "auth = TokenAuthentication(\n", + " token = \"XXXXX\",\n", + " server = \"XXXXX\",\n", + " skip_tls=False\n", + ")\n", + "auth.login()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "bc27f84c", + "metadata": {}, + "source": [ + "This time, we are working in a cloud environment, and our OpenShift cluster does not have the resources needed for our desired workloads. We will use InstaScale to dynamically scale-up guaranteed resources based on our request (that will also automatically scale-down when we are finished working):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0f4bc870-091f-4e11-9642-cba145710159", + "metadata": {}, + "outputs": [], + "source": [ + "# Create and configure our cluster object (and appwrapper)\n", + "cluster = Cluster(ClusterConfiguration(\n", + " name='instascaletest',\n", + " namespace='default',\n", + " min_worker=2,\n", + " max_worker=2,\n", + " min_cpus=2,\n", + " max_cpus=2,\n", + " min_memory=8,\n", + " max_memory=8,\n", + " gpu=1,\n", + " instascale=True, # InstaScale now enabled, will scale OCP cluster to guarantee resource request\n", + " machine_types=[\"m5.xlarge\", \"g4dn.xlarge\"] # Head, worker AWS machine types desired\n", + "))" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "12eef53c", + "metadata": {}, + "source": [ + "Same as last time, we will bring the cluster up, wait for it to be ready, and confirm that the specs are as-requested:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f0884bbc-c224-4ca0-98a0-02dfa09c2200", + "metadata": {}, + "outputs": [], + "source": [ + "# Bring up the cluster\n", + "cluster.up()\n", + "cluster.wait_ready()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "6abfe904", + "metadata": {}, + "source": [ + "While the resources are being scaled, we can also go into the console and take a look at the InstaScale logs, as well as the new machines/nodes spinning up.\n", + "\n", + "Once the cluster is ready, we can confirm the specs:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7fd45bc5-03c0-4ae5-9ec5-dd1c30f1a084", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.details()" + ] + }, + { + "cell_type": "markdown", + "id": "5af8cd32", + "metadata": {}, + "source": [ + "Finally, we bring our resource cluster down and release/terminate the associated resources, bringing everything back to the way it was before our cluster was brought up." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5f36db0f-31f6-4373-9503-dc3c1c4c3f57", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.down()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "c883caea", + "metadata": {}, + "source": [ + "Once again, we can look at the machines/nodes and see that everything has been successfully scaled down!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0d41b90e", + "metadata": {}, + "outputs": [], + "source": [ + "auth.logout()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.9.13" + }, + "vscode": { + "interpreter": { + "hash": "f9f85f796d01129d0dd105a088854619f454435301f6ffec2fea96ecbd9be4ac" + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/demo-notebooks/guided-demos/2_basic_jobs.ipynb b/demo-notebooks/guided-demos/2_basic_jobs.ipynb new file mode 100644 index 000000000..37560b172 --- /dev/null +++ b/demo-notebooks/guided-demos/2_basic_jobs.ipynb @@ -0,0 +1,298 @@ +{ + "cells": [ + { + "attachments": {}, + "cell_type": "markdown", + "id": "464af595", + "metadata": {}, + "source": [ + "In this third notebook, we will go over the basics of submitting jobs via the SDK, either to a Ray cluster or directly to MCAD." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b55bc3ea-4ce3-49bf-bb1f-e209de8ca47a", + "metadata": {}, + "outputs": [], + "source": [ + "# Import pieces from codeflare-sdk\n", + "from codeflare_sdk.cluster.cluster import Cluster, ClusterConfiguration\n", + "from codeflare_sdk.cluster.auth import TokenAuthentication" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "614daa0c", + "metadata": {}, + "outputs": [], + "source": [ + "# Create authentication object for oc user permissions\n", + "auth = TokenAuthentication(\n", + " token = \"XXXXX\",\n", + " server = \"XXXXX\",\n", + " skip_tls=False\n", + ")\n", + "auth.login()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "bc27f84c", + "metadata": {}, + "source": [ + "Let's start by running through the same cluster setup as before:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0f4bc870-091f-4e11-9642-cba145710159", + "metadata": {}, + "outputs": [], + "source": [ + "# Create and configure our cluster object (and appwrapper)\n", + "cluster = Cluster(ClusterConfiguration(\n", + " name='jobtest',\n", + " namespace='default',\n", + " min_worker=2,\n", + " max_worker=2,\n", + " min_cpus=1,\n", + " max_cpus=1,\n", + " min_memory=4,\n", + " max_memory=4,\n", + " gpu=0,\n", + " instascale=False\n", + "))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f0884bbc-c224-4ca0-98a0-02dfa09c2200", + "metadata": {}, + "outputs": [], + "source": [ + "# Bring up the cluster\n", + "cluster.up()\n", + "cluster.wait_ready()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "df71c1ed", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.details()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "33663f47", + "metadata": {}, + "source": [ + "This time, however, we are going to use the CodeFlare SDK to submit batch jobs via TorchX, either to the Ray cluster we have just brought up, or directly to MCAD." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c7b4f232", + "metadata": {}, + "outputs": [], + "source": [ + "from codeflare_sdk.job.jobs import DDPJobDefinition" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "83d77b74", + "metadata": {}, + "source": [ + "First, let's begin by submitting to Ray, training a basic NN on the MNIST dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8c2c5138", + "metadata": {}, + "outputs": [], + "source": [ + "jobdef = DDPJobDefinition(\n", + " name=\"mnisttest\",\n", + " script=\"mnist.py\",\n", + " scheduler_args={\"requirements\": \"requirements.txt\"}\n", + ")\n", + "job = jobdef.submit(cluster)" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "5b9ae53a", + "metadata": {}, + "source": [ + "Now we can take a look at the status of our submitted job, as well as the logs:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "6e36c3d9", + "metadata": {}, + "outputs": [], + "source": [ + "job.status()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "834cfb5c", + "metadata": {}, + "outputs": [], + "source": [ + "job.logs()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "5af8cd32", + "metadata": {}, + "source": [ + "Once complete, we can bring our Ray cluster down and clean up:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5f36db0f-31f6-4373-9503-dc3c1c4c3f57", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.down()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "31096641", + "metadata": {}, + "source": [ + "Now, an alternative option for job submission is to submit directly to MCAD, which will schedule pods to run the job with requested resources:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "496139cc", + "metadata": {}, + "outputs": [], + "source": [ + "jobdef = DDPJobDefinition(\n", + " name=\"mnistjob\",\n", + " script=\"mnist.py\",\n", + " scheduler_args={\"namespace\": \"default\"},\n", + " j=\"1x1\",\n", + " gpu=0,\n", + " cpu=1,\n", + " memMB=8000,\n", + " image=\"quay.io/project-codeflare/mnist-job-test:v0.0.1\"\n", + ")\n", + "job = jobdef.submit()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "0837e43b", + "metadata": {}, + "source": [ + "Once again, we can look at job status and logs:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3d18d42c", + "metadata": {}, + "outputs": [], + "source": [ + "job.status()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "36d7ea97", + "metadata": {}, + "outputs": [], + "source": [ + "job.logs()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "aebf376a", + "metadata": {}, + "source": [ + "This time, once the pods complete, we can clean them up alongside any other associated resources. The following command can also be used to delete jobs early for both Ray and MCAD submission:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "ebbb0674", + "metadata": {}, + "outputs": [], + "source": [ + "job.cancel()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0d41b90e", + "metadata": {}, + "outputs": [], + "source": [ + "auth.logout()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.9.13" + }, + "vscode": { + "interpreter": { + "hash": "f9f85f796d01129d0dd105a088854619f454435301f6ffec2fea96ecbd9be4ac" + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/demo-notebooks/guided-demos/3_basic_interactive.ipynb b/demo-notebooks/guided-demos/3_basic_interactive.ipynb new file mode 100644 index 000000000..4d5031984 --- /dev/null +++ b/demo-notebooks/guided-demos/3_basic_interactive.ipynb @@ -0,0 +1,302 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "bbc21043", + "metadata": {}, + "source": [ + "In this fourth and final notebook, we will go over how to leverage the SDK to directly work interactively with a Ray cluster during development." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "b55bc3ea-4ce3-49bf-bb1f-e209de8ca47a", + "metadata": {}, + "outputs": [], + "source": [ + "# Import pieces from codeflare-sdk\n", + "from codeflare_sdk.cluster.cluster import Cluster, ClusterConfiguration\n", + "from codeflare_sdk.cluster.auth import TokenAuthentication" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "614daa0c", + "metadata": {}, + "outputs": [], + "source": [ + "# Create authentication object for oc user permissions\n", + "auth = TokenAuthentication(\n", + " token = \"XXXXX\",\n", + " server = \"XXXXX\",\n", + " skip_tls=False\n", + ")\n", + "auth.login()" + ] + }, + { + "cell_type": "markdown", + "id": "bc27f84c", + "metadata": {}, + "source": [ + "Once again, let's start by running through the same cluster setup as before:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0f4bc870-091f-4e11-9642-cba145710159", + "metadata": {}, + "outputs": [], + "source": [ + "# Create and configure our cluster object (and appwrapper)\n", + "cluster = Cluster(ClusterConfiguration(\n", + " name='interactivetest',\n", + " namespace='default',\n", + " min_worker=2,\n", + " max_worker=2,\n", + " min_cpus=2,\n", + " max_cpus=2,\n", + " min_memory=8,\n", + " max_memory=8,\n", + " gpu=1,\n", + " instascale=True,\n", + " machine_types=[\"m5.xlarge\", \"g4dn.xlarge\"]\n", + " \n", + "))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "f0884bbc-c224-4ca0-98a0-02dfa09c2200", + "metadata": {}, + "outputs": [], + "source": [ + "# Bring up the cluster\n", + "cluster.up()\n", + "cluster.wait_ready()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "df71c1ed", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.details()" + ] + }, + { + "cell_type": "markdown", + "id": "33663f47", + "metadata": {}, + "source": [ + "This time we will demonstrate another potential method of use: working with the Ray cluster interactively.\n", + "\n", + "Using the SDK, we can get both the Ray cluster URI and dashboard URI:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c1719bca", + "metadata": {}, + "outputs": [], + "source": [ + "ray_dashboard_uri = cluster.cluster_dashboard_uri()\n", + "ray_cluster_uri = cluster.cluster_uri()\n", + "print(ray_dashboard_uri)\n", + "print(ray_cluster_uri)" + ] + }, + { + "cell_type": "markdown", + "id": "2a2aca6a", + "metadata": {}, + "source": [ + "Now we can connect directly to our Ray cluster via the Ray python client:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "300146dc", + "metadata": {}, + "outputs": [], + "source": [ + "#before proceeding make sure the cluster exists and the uri is not empty\n", + "assert ray_cluster_uri, \"Ray cluster needs to be started and set before proceeding\"\n", + "\n", + "import ray\n", + "from ray.air.config import ScalingConfig\n", + "\n", + "# reset the ray context in case there's already one. \n", + "ray.shutdown()\n", + "# establish connection to ray cluster\n", + "\n", + "#install additionall libraries that will be required for model training\n", + "runtime_env = {\"pip\": [\"transformers\", \"datasets\", \"evaluate\", \"pyarrow<7.0.0\"]}\n", + "\n", + "ray.init(address=f'{ray_cluster_uri}', runtime_env=runtime_env)\n", + "\n", + "print(\"Ray cluster is up and running: \", ray.is_initialized())" + ] + }, + { + "cell_type": "markdown", + "id": "9711030b", + "metadata": {}, + "source": [ + "Now that we are connected (and have passed in some package requirements), let's try writing some training code for a DistilBERT transformer model via HuggingFace (using IMDB dataset):" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1b36e0d9", + "metadata": {}, + "outputs": [], + "source": [ + "@ray.remote\n", + "def train_fn():\n", + " from datasets import load_dataset\n", + " import transformers\n", + " from transformers import AutoTokenizer, TrainingArguments\n", + " from transformers import AutoModelForSequenceClassification\n", + " import numpy as np\n", + " from datasets import load_metric\n", + " import ray\n", + " from ray import tune\n", + " from ray.train.huggingface import HuggingFaceTrainer\n", + "\n", + " dataset = load_dataset(\"imdb\")\n", + " tokenizer = AutoTokenizer.from_pretrained(\"distilbert-base-uncased\")\n", + "\n", + " def tokenize_function(examples):\n", + " return tokenizer(examples[\"text\"], padding=\"max_length\", truncation=True)\n", + "\n", + " tokenized_datasets = dataset.map(tokenize_function, batched=True)\n", + "\n", + " #using a fraction of dataset but you can run with the full dataset\n", + " small_train_dataset = tokenized_datasets[\"train\"].shuffle(seed=42).select(range(100))\n", + " small_eval_dataset = tokenized_datasets[\"test\"].shuffle(seed=42).select(range(100))\n", + "\n", + " print(f\"len of train {small_train_dataset} and test {small_eval_dataset}\")\n", + "\n", + " ray_train_ds = ray.data.from_huggingface(small_train_dataset)\n", + " ray_evaluation_ds = ray.data.from_huggingface(small_eval_dataset)\n", + "\n", + " def compute_metrics(eval_pred):\n", + " metric = load_metric(\"accuracy\")\n", + " logits, labels = eval_pred\n", + " predictions = np.argmax(logits, axis=-1)\n", + " return metric.compute(predictions=predictions, references=labels)\n", + "\n", + " def trainer_init_per_worker(train_dataset, eval_dataset, **config):\n", + " model = AutoModelForSequenceClassification.from_pretrained(\"distilbert-base-uncased\", num_labels=2)\n", + "\n", + " training_args = TrainingArguments(\"/tmp/hf_imdb/test\", eval_steps=1, disable_tqdm=True, \n", + " num_train_epochs=1, skip_memory_metrics=True,\n", + " learning_rate=2e-5,\n", + " per_device_train_batch_size=16,\n", + " per_device_eval_batch_size=16, \n", + " weight_decay=0.01,)\n", + " return transformers.Trainer(\n", + " model=model,\n", + " args=training_args,\n", + " train_dataset=train_dataset,\n", + " eval_dataset=eval_dataset,\n", + " compute_metrics=compute_metrics\n", + " )\n", + "\n", + " scaling_config = ScalingConfig(num_workers=2, use_gpu=True) #num workers is the number of gpus\n", + "\n", + " # we are using the ray native HuggingFaceTrainer, but you can swap out to use non ray Huggingface Trainer. Both have the same method signature. \n", + " # the ray native HFTrainer has built in support for scaling to multiple GPUs\n", + " trainer = HuggingFaceTrainer(\n", + " trainer_init_per_worker=trainer_init_per_worker,\n", + " scaling_config=scaling_config,\n", + " datasets={\"train\": ray_train_ds, \"evaluation\": ray_evaluation_ds},\n", + " )\n", + " result = trainer.fit()" + ] + }, + { + "cell_type": "markdown", + "id": "d4d8fd65", + "metadata": {}, + "source": [ + "Once we want to test our code out, we can run the training function we defined above remotely on our Ray cluster:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5901d958", + "metadata": {}, + "outputs": [], + "source": [ + "#call the above cell as a remote ray function\n", + "ray.get(train_fn.remote())" + ] + }, + { + "cell_type": "markdown", + "id": "5af8cd32", + "metadata": {}, + "source": [ + "Once complete, we can bring our Ray cluster down and clean up:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5f36db0f-31f6-4373-9503-dc3c1c4c3f57", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.down()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0d41b90e", + "metadata": {}, + "outputs": [], + "source": [ + "auth.logout()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.9.13" + }, + "vscode": { + "interpreter": { + "hash": "f9f85f796d01129d0dd105a088854619f454435301f6ffec2fea96ecbd9be4ac" + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/demo-notebooks/guided-demos/4_gpt.ipynb b/demo-notebooks/guided-demos/4_gpt.ipynb new file mode 100644 index 000000000..0132bd4a8 --- /dev/null +++ b/demo-notebooks/guided-demos/4_gpt.ipynb @@ -0,0 +1,191 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": null, + "id": "b6c05b69-4ce8-45ef-82d3-bacb2491bee8", + "metadata": {}, + "outputs": [], + "source": [ + "# Import pieces from codeflare-sdk\n", + "from codeflare_sdk.cluster.cluster import Cluster, ClusterConfiguration\n", + "from codeflare_sdk.cluster.auth import TokenAuthentication" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "32f99bbd-9903-4d38-a4f2-223dec684ae2", + "metadata": {}, + "outputs": [], + "source": [ + "# Create authentication object for oc user permissions\n", + "auth = TokenAuthentication(\n", + " token = \"XXXXX\",\n", + " server = \"XXXXX\",\n", + " skip_tls=False\n", + ")\n", + "auth.login()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "3f32119a-c4ee-4163-b103-d9ca3bddbdb5", + "metadata": {}, + "outputs": [], + "source": [ + "cluster = Cluster(ClusterConfiguration(\n", + " name='gptfttest',\n", + " namespace='default',\n", + " min_worker=2,\n", + " max_worker=2,\n", + " min_cpus=2,\n", + " max_cpus=2,\n", + " min_memory=8,\n", + " max_memory=8,\n", + " gpu=1,\n", + " instascale=True,\n", + " machine_types=[\"m5.xlarge\", \"g4dn.xlarge\"],\n", + "))" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "107c8277-3b3b-4238-a786-a391a662fd7c", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.up()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "730f66ce-adaa-4709-b9cf-22417847e059", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.wait_ready()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "48fac218-2f22-428b-9228-137a4bb0e666", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.details()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9ed5bd75-4230-4c7c-a9e2-0f247890e62a", + "metadata": {}, + "outputs": [], + "source": [ + "from codeflare_sdk.job.jobs import DDPJobDefinition" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "611d203a-35aa-4357-a748-1d01b022fcdb", + "metadata": {}, + "outputs": [], + "source": [ + "arg_list = [\n", + " \"--model_name_or_path\", \"gpt2\",\n", + " \"--dataset_name\", \"wikitext\",\n", + " \"--dataset_config_name\", \"wikitext-2-raw-v1\",\n", + " \"--per_device_train_batch_size\", \"2\",\n", + " \"--per_device_eval_batch_size\", \"2\",\n", + " \"--do_train\",\n", + " \"--do_eval\",\n", + " \"--output_dir\", \"/tmp/test-clm\",\n", + " \"--overwrite_output_dir\"\n", + "]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8ac7c34f-e227-44c2-a4b1-a57c853ac3a7", + "metadata": {}, + "outputs": [], + "source": [ + "jobdef = DDPJobDefinition(\n", + " name=\"gpttest\",\n", + " script=\"gpt_og.py\",\n", + " script_args=arg_list,\n", + " scheduler_args={\"requirements\": \"requirements_gpt.txt\"}\n", + ")\n", + "job = jobdef.submit(cluster)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1680d287-de46-45f8-b95a-02ba3c83912c", + "metadata": {}, + "outputs": [], + "source": [ + "job.status()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "d25d6198-9941-47e8-857f-9811830cc854", + "metadata": {}, + "outputs": [], + "source": [ + "job.logs()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "beb1a6b9-d9b3-49b7-b036-09f1d3569b59", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.down()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8398d977-db24-46d0-a7d2-b4e9197808d7", + "metadata": {}, + "outputs": [], + "source": [ + "auth.logout()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.13" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/demo-notebooks/guided-demos/gpt_og.py b/demo-notebooks/guided-demos/gpt_og.py new file mode 100644 index 000000000..d69e41fcb --- /dev/null +++ b/demo-notebooks/guided-demos/gpt_og.py @@ -0,0 +1,728 @@ +#!/usr/bin/env python +# coding=utf-8 +# Copyright 2020 The HuggingFace Inc. team. All rights reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +""" +Fine-tuning the library models for causal language modeling (GPT, GPT-2, CTRL, ...) on a text file or a dataset. + +Here is the full list of checkpoints on the hub that can be fine-tuned by this script: +https://huggingface.co/models?filter=text-generation +""" +# You can also adapt this script on your own causal language modeling task. Pointers for this are left as comments. + +import subprocess + +subprocess.run(["pip", "uninstall", "protobuf"]) +subprocess.run( + [ + "pip", + "install", + "--upgrade", + "--target=/home/ray/workspace", + "-r", + "requirements.txt", + ] +) + +import logging +import math +import os +import sys +from dataclasses import dataclass, field +from itertools import chain +from typing import Optional + +import datasets +import evaluate +import torch +from datasets import load_dataset + +import transformers +from transformers import ( + CONFIG_MAPPING, + MODEL_FOR_CAUSAL_LM_MAPPING, + AutoConfig, + AutoModelForCausalLM, + AutoTokenizer, + HfArgumentParser, + Trainer, + TrainingArguments, + default_data_collator, + is_torch_tpu_available, + set_seed, +) +from transformers.testing_utils import CaptureLogger +from transformers.trainer_utils import get_last_checkpoint +from transformers.utils import check_min_version, send_example_telemetry +from transformers.utils.versions import require_version + + +# Will error if the minimal version of Transformers is not installed. Remove at your own risks. +# check_min_version("4.29.0.dev0") + +require_version( + "datasets>=1.8.0", + "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt", +) + +logger = logging.getLogger(__name__) + + +MODEL_CONFIG_CLASSES = list(MODEL_FOR_CAUSAL_LM_MAPPING.keys()) +MODEL_TYPES = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES) + + +@dataclass +class ModelArguments: + """ + Arguments pertaining to which model/config/tokenizer we are going to fine-tune, or train from scratch. + """ + + model_name_or_path: Optional[str] = field( + default=None, + metadata={ + "help": ( + "The model checkpoint for weights initialization.Don't set if you want to train a model from scratch." + ) + }, + ) + model_type: Optional[str] = field( + default=None, + metadata={ + "help": "If training from scratch, pass a model type from the list: " + + ", ".join(MODEL_TYPES) + }, + ) + config_overrides: Optional[str] = field( + default=None, + metadata={ + "help": ( + "Override some existing default config settings when a model is trained from scratch. Example: " + "n_embd=10,resid_pdrop=0.2,scale_attn_weights=false,summary_type=cls_index" + ) + }, + ) + config_name: Optional[str] = field( + default=None, + metadata={ + "help": "Pretrained config name or path if not the same as model_name" + }, + ) + tokenizer_name: Optional[str] = field( + default=None, + metadata={ + "help": "Pretrained tokenizer name or path if not the same as model_name" + }, + ) + cache_dir: Optional[str] = field( + default=None, + metadata={ + "help": "Where do you want to store the pretrained models downloaded from huggingface.co" + }, + ) + use_fast_tokenizer: bool = field( + default=True, + metadata={ + "help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not." + }, + ) + model_revision: str = field( + default="main", + metadata={ + "help": "The specific model version to use (can be a branch name, tag name or commit id)." + }, + ) + use_auth_token: bool = field( + default=False, + metadata={ + "help": ( + "Will use the token generated when running `huggingface-cli login` (necessary to use this script " + "with private models)." + ) + }, + ) + torch_dtype: Optional[str] = field( + default=None, + metadata={ + "help": ( + "Override the default `torch.dtype` and load the model under this dtype. If `auto` is passed, the " + "dtype will be automatically derived from the model's weights." + ), + "choices": ["auto", "bfloat16", "float16", "float32"], + }, + ) + low_cpu_mem_usage: bool = field( + default=False, + metadata={ + "help": ( + "It is an option to create the model as an empty shell, then only materialize its parameters when the pretrained weights are loaded." + "set True will benefit LLM loading time and RAM consumption." + ) + }, + ) + + def __post_init__(self): + if self.config_overrides is not None and ( + self.config_name is not None or self.model_name_or_path is not None + ): + raise ValueError( + "--config_overrides can't be used in combination with --config_name or --model_name_or_path" + ) + + +@dataclass +class DataTrainingArguments: + """ + Arguments pertaining to what data we are going to input our model for training and eval. + """ + + dataset_name: Optional[str] = field( + default=None, + metadata={"help": "The name of the dataset to use (via the datasets library)."}, + ) + dataset_config_name: Optional[str] = field( + default=None, + metadata={ + "help": "The configuration name of the dataset to use (via the datasets library)." + }, + ) + train_file: Optional[str] = field( + default=None, metadata={"help": "The input training data file (a text file)."} + ) + validation_file: Optional[str] = field( + default=None, + metadata={ + "help": "An optional input evaluation data file to evaluate the perplexity on (a text file)." + }, + ) + max_train_samples: Optional[int] = field( + default=None, + metadata={ + "help": ( + "For debugging purposes or quicker training, truncate the number of training examples to this " + "value if set." + ) + }, + ) + max_eval_samples: Optional[int] = field( + default=None, + metadata={ + "help": ( + "For debugging purposes or quicker training, truncate the number of evaluation examples to this " + "value if set." + ) + }, + ) + streaming: bool = field(default=False, metadata={"help": "Enable streaming mode"}) + block_size: Optional[int] = field( + default=None, + metadata={ + "help": ( + "Optional input sequence length after tokenization. " + "The training dataset will be truncated in block of this size for training. " + "Default to the model max input length for single sentence inputs (take into account special tokens)." + ) + }, + ) + overwrite_cache: bool = field( + default=False, + metadata={"help": "Overwrite the cached training and evaluation sets"}, + ) + validation_split_percentage: Optional[int] = field( + default=5, + metadata={ + "help": "The percentage of the train set used as validation set in case there's no validation split" + }, + ) + preprocessing_num_workers: Optional[int] = field( + default=None, + metadata={"help": "The number of processes to use for the preprocessing."}, + ) + keep_linebreaks: bool = field( + default=True, + metadata={"help": "Whether to keep line breaks when using TXT files or not."}, + ) + + def __post_init__(self): + if self.streaming: + require_version( + "datasets>=2.0.0", "The streaming feature requires `datasets>=2.0.0`" + ) + + if ( + self.dataset_name is None + and self.train_file is None + and self.validation_file is None + ): + raise ValueError( + "Need either a dataset name or a training/validation file." + ) + else: + if self.train_file is not None: + extension = self.train_file.split(".")[-1] + assert extension in [ + "csv", + "json", + "txt", + ], "`train_file` should be a csv, a json or a txt file." + if self.validation_file is not None: + extension = self.validation_file.split(".")[-1] + assert extension in [ + "csv", + "json", + "txt", + ], "`validation_file` should be a csv, a json or a txt file." + + +def main(): + # See all possible arguments in src/transformers/training_args.py + # or by passing the --help flag to this script. + # We now keep distinct sets of args, for a cleaner separation of concerns. + + parser = HfArgumentParser( + (ModelArguments, DataTrainingArguments, TrainingArguments) + ) + if len(sys.argv) == 2 and sys.argv[1].endswith(".json"): + # If we pass only one argument to the script and it's the path to a json file, + # let's parse it to get our arguments. + model_args, data_args, training_args = parser.parse_json_file( + json_file=os.path.abspath(sys.argv[1]) + ) + else: + model_args, data_args, training_args = parser.parse_args_into_dataclasses() + + # Sending telemetry. Tracking the example usage helps us better allocate resources to maintain them. The + # information sent is the one passed as arguments along with your Python/PyTorch versions. + send_example_telemetry("run_clm", model_args, data_args) + + # Setup logging + logging.basicConfig( + format="%(asctime)s - %(levelname)s - %(name)s - %(message)s", + datefmt="%m/%d/%Y %H:%M:%S", + handlers=[logging.StreamHandler(sys.stdout)], + ) + + if training_args.should_log: + # The default of training_args.log_level is passive, so we set log level at info here to have that default. + transformers.utils.logging.set_verbosity_info() + + log_level = training_args.get_process_log_level() + logger.setLevel(log_level) + datasets.utils.logging.set_verbosity(log_level) + transformers.utils.logging.set_verbosity(log_level) + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() + + # Log on each process the small summary: + logger.warning( + f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}" + + f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}" + ) + logger.info(f"Training/evaluation parameters {training_args}") + + # Detecting last checkpoint. + last_checkpoint = None + if ( + os.path.isdir(training_args.output_dir) + and training_args.do_train + and not training_args.overwrite_output_dir + ): + last_checkpoint = get_last_checkpoint(training_args.output_dir) + if last_checkpoint is None and len(os.listdir(training_args.output_dir)) > 0: + raise ValueError( + f"Output directory ({training_args.output_dir}) already exists and is not empty. " + "Use --overwrite_output_dir to overcome." + ) + elif ( + last_checkpoint is not None and training_args.resume_from_checkpoint is None + ): + logger.info( + f"Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change " + "the `--output_dir` or add `--overwrite_output_dir` to train from scratch." + ) + + # Set seed before initializing model. + set_seed(training_args.seed) + + # Get the datasets: you can either provide your own CSV/JSON/TXT training and evaluation files (see below) + # or just provide the name of one of the public datasets available on the hub at https://huggingface.co/datasets/ + # (the dataset will be downloaded automatically from the datasets Hub). + # + # For CSV/JSON files, this script will use the column called 'text' or the first column if no column called + # 'text' is found. You can easily tweak this behavior (see below). + # + # In distributed training, the load_dataset function guarantee that only one local process can concurrently + # download the dataset. + if data_args.dataset_name is not None: + # Downloading and loading a dataset from the hub. + raw_datasets = load_dataset( + data_args.dataset_name, + data_args.dataset_config_name, + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + streaming=data_args.streaming, + ) + if "validation" not in raw_datasets.keys(): + raw_datasets["validation"] = load_dataset( + data_args.dataset_name, + data_args.dataset_config_name, + split=f"train[:{data_args.validation_split_percentage}%]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + streaming=data_args.streaming, + ) + raw_datasets["train"] = load_dataset( + data_args.dataset_name, + data_args.dataset_config_name, + split=f"train[{data_args.validation_split_percentage}%:]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + streaming=data_args.streaming, + ) + else: + data_files = {} + dataset_args = {} + if data_args.train_file is not None: + data_files["train"] = data_args.train_file + if data_args.validation_file is not None: + data_files["validation"] = data_args.validation_file + extension = ( + data_args.train_file.split(".")[-1] + if data_args.train_file is not None + else data_args.validation_file.split(".")[-1] + ) + if extension == "txt": + extension = "text" + dataset_args["keep_linebreaks"] = data_args.keep_linebreaks + raw_datasets = load_dataset( + extension, + data_files=data_files, + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + **dataset_args, + ) + # If no validation data is there, validation_split_percentage will be used to divide the dataset. + if "validation" not in raw_datasets.keys(): + raw_datasets["validation"] = load_dataset( + extension, + data_files=data_files, + split=f"train[:{data_args.validation_split_percentage}%]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + **dataset_args, + ) + raw_datasets["train"] = load_dataset( + extension, + data_files=data_files, + split=f"train[{data_args.validation_split_percentage}%:]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + **dataset_args, + ) + + # See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at + # https://huggingface.co/docs/datasets/loading_datasets.html. + + # Load pretrained model and tokenizer + # + # Distributed training: + # The .from_pretrained methods guarantee that only one local process can concurrently + # download model & vocab. + + config_kwargs = { + "cache_dir": model_args.cache_dir, + "revision": model_args.model_revision, + "use_auth_token": True if model_args.use_auth_token else None, + } + if model_args.config_name: + config = AutoConfig.from_pretrained(model_args.config_name, **config_kwargs) + elif model_args.model_name_or_path: + config = AutoConfig.from_pretrained( + model_args.model_name_or_path, **config_kwargs + ) + else: + config = CONFIG_MAPPING[model_args.model_type]() + logger.warning("You are instantiating a new config instance from scratch.") + if model_args.config_overrides is not None: + logger.info(f"Overriding config: {model_args.config_overrides}") + config.update_from_string(model_args.config_overrides) + logger.info(f"New config: {config}") + + tokenizer_kwargs = { + "cache_dir": model_args.cache_dir, + "use_fast": model_args.use_fast_tokenizer, + "revision": model_args.model_revision, + "use_auth_token": True if model_args.use_auth_token else None, + } + if model_args.tokenizer_name: + tokenizer = AutoTokenizer.from_pretrained( + model_args.tokenizer_name, **tokenizer_kwargs + ) + elif model_args.model_name_or_path: + tokenizer = AutoTokenizer.from_pretrained( + model_args.model_name_or_path, **tokenizer_kwargs + ) + else: + raise ValueError( + "You are instantiating a new tokenizer from scratch. This is not supported by this script." + "You can do it from another script, save it, and load it from here, using --tokenizer_name." + ) + + if model_args.model_name_or_path: + torch_dtype = ( + model_args.torch_dtype + if model_args.torch_dtype in ["auto", None] + else getattr(torch, model_args.torch_dtype) + ) + model = AutoModelForCausalLM.from_pretrained( + model_args.model_name_or_path, + from_tf=bool(".ckpt" in model_args.model_name_or_path), + config=config, + cache_dir=model_args.cache_dir, + revision=model_args.model_revision, + use_auth_token=True if model_args.use_auth_token else None, + torch_dtype=torch_dtype, + low_cpu_mem_usage=model_args.low_cpu_mem_usage, + ) + else: + model = AutoModelForCausalLM.from_config(config) + n_params = sum({p.data_ptr(): p.numel() for p in model.parameters()}.values()) + logger.info( + f"Training new model from scratch - Total size={n_params/2**20:.2f}M params" + ) + + # We resize the embeddings only when necessary to avoid index errors. If you are creating a model from scratch + # on a small vocab and want a smaller embedding size, remove this test. + embedding_size = model.get_input_embeddings().weight.shape[0] + if len(tokenizer) > embedding_size: + model.resize_token_embeddings(len(tokenizer)) + + # Preprocessing the datasets. + # First we tokenize all the texts. + if training_args.do_train: + column_names = list(raw_datasets["train"].features) + else: + column_names = list(raw_datasets["validation"].features) + text_column_name = "text" if "text" in column_names else column_names[0] + + # since this will be pickled to avoid _LazyModule error in Hasher force logger loading before tokenize_function + tok_logger = transformers.utils.logging.get_logger( + "transformers.tokenization_utils_base" + ) + + def tokenize_function(examples): + with CaptureLogger(tok_logger) as cl: + output = tokenizer(examples[text_column_name]) + # clm input could be much much longer than block_size + if "Token indices sequence length is longer than the" in cl.out: + tok_logger.warning( + "^^^^^^^^^^^^^^^^ Please ignore the warning above - this long input will be chunked into smaller bits" + " before being passed to the model." + ) + return output + + with training_args.main_process_first(desc="dataset map tokenization"): + if not data_args.streaming: + tokenized_datasets = raw_datasets.map( + tokenize_function, + batched=True, + num_proc=data_args.preprocessing_num_workers, + remove_columns=column_names, + load_from_cache_file=not data_args.overwrite_cache, + desc="Running tokenizer on dataset", + ) + else: + tokenized_datasets = raw_datasets.map( + tokenize_function, + batched=True, + remove_columns=column_names, + ) + + if data_args.block_size is None: + block_size = tokenizer.model_max_length + if block_size > 1024: + logger.warning( + "The chosen tokenizer supports a `model_max_length` that is longer than the default `block_size` value" + " of 1024. If you would like to use a longer `block_size` up to `tokenizer.model_max_length` you can" + " override this default with `--block_size xxx`." + ) + block_size = 1024 + else: + if data_args.block_size > tokenizer.model_max_length: + logger.warning( + f"The block_size passed ({data_args.block_size}) is larger than the maximum length for the model" + f"({tokenizer.model_max_length}). Using block_size={tokenizer.model_max_length}." + ) + block_size = min(data_args.block_size, tokenizer.model_max_length) + + # Main data processing function that will concatenate all texts from our dataset and generate chunks of block_size. + def group_texts(examples): + # Concatenate all texts. + concatenated_examples = {k: list(chain(*examples[k])) for k in examples.keys()} + total_length = len(concatenated_examples[list(examples.keys())[0]]) + # We drop the small remainder, we could add padding if the model supported it instead of this drop, you can + # customize this part to your needs. + if total_length >= block_size: + total_length = (total_length // block_size) * block_size + # Split by chunks of max_len. + result = { + k: [t[i : i + block_size] for i in range(0, total_length, block_size)] + for k, t in concatenated_examples.items() + } + result["labels"] = result["input_ids"].copy() + return result + + # Note that with `batched=True`, this map processes 1,000 texts together, so group_texts throws away a remainder + # for each of those groups of 1,000 texts. You can adjust that batch_size here but a higher value might be slower + # to preprocess. + # + # To speed up this part, we use multiprocessing. See the documentation of the map method for more information: + # https://huggingface.co/docs/datasets/package_reference/main_classes.html#datasets.Dataset.map + + with training_args.main_process_first(desc="grouping texts together"): + if not data_args.streaming: + lm_datasets = tokenized_datasets.map( + group_texts, + batched=True, + num_proc=data_args.preprocessing_num_workers, + load_from_cache_file=not data_args.overwrite_cache, + desc=f"Grouping texts in chunks of {block_size}", + ) + else: + lm_datasets = tokenized_datasets.map( + group_texts, + batched=True, + ) + + if training_args.do_train: + if "train" not in tokenized_datasets: + raise ValueError("--do_train requires a train dataset") + train_dataset = lm_datasets["train"] + if data_args.max_train_samples is not None: + max_train_samples = min(len(train_dataset), data_args.max_train_samples) + train_dataset = train_dataset.select(range(max_train_samples)) + + if training_args.do_eval: + if "validation" not in tokenized_datasets: + raise ValueError("--do_eval requires a validation dataset") + eval_dataset = lm_datasets["validation"] + if data_args.max_eval_samples is not None: + max_eval_samples = min(len(eval_dataset), data_args.max_eval_samples) + eval_dataset = eval_dataset.select(range(max_eval_samples)) + + def preprocess_logits_for_metrics(logits, labels): + if isinstance(logits, tuple): + # Depending on the model and config, logits may contain extra tensors, + # like past_key_values, but logits always come first + logits = logits[0] + return logits.argmax(dim=-1) + + metric = evaluate.load("accuracy") + + def compute_metrics(eval_preds): + preds, labels = eval_preds + # preds have the same shape as the labels, after the argmax(-1) has been calculated + # by preprocess_logits_for_metrics but we need to shift the labels + labels = labels[:, 1:].reshape(-1) + preds = preds[:, :-1].reshape(-1) + return metric.compute(predictions=preds, references=labels) + + # Initialize our Trainer + trainer = Trainer( + model=model, + args=training_args, + train_dataset=train_dataset if training_args.do_train else None, + eval_dataset=eval_dataset if training_args.do_eval else None, + tokenizer=tokenizer, + # Data collator will default to DataCollatorWithPadding, so we change it. + data_collator=default_data_collator, + compute_metrics=compute_metrics + if training_args.do_eval and not is_torch_tpu_available() + else None, + preprocess_logits_for_metrics=preprocess_logits_for_metrics + if training_args.do_eval and not is_torch_tpu_available() + else None, + ) + + # Training + if training_args.do_train: + checkpoint = None + if training_args.resume_from_checkpoint is not None: + checkpoint = training_args.resume_from_checkpoint + elif last_checkpoint is not None: + checkpoint = last_checkpoint + train_result = trainer.train(resume_from_checkpoint=checkpoint) + trainer.save_model() # Saves the tokenizer too for easy upload + + metrics = train_result.metrics + + max_train_samples = ( + data_args.max_train_samples + if data_args.max_train_samples is not None + else len(train_dataset) + ) + metrics["train_samples"] = min(max_train_samples, len(train_dataset)) + + trainer.log_metrics("train", metrics) + trainer.save_metrics("train", metrics) + trainer.save_state() + + # Evaluation + if training_args.do_eval: + logger.info("*** Evaluate ***") + + metrics = trainer.evaluate() + + max_eval_samples = ( + data_args.max_eval_samples + if data_args.max_eval_samples is not None + else len(eval_dataset) + ) + metrics["eval_samples"] = min(max_eval_samples, len(eval_dataset)) + try: + perplexity = math.exp(metrics["eval_loss"]) + except OverflowError: + perplexity = float("inf") + metrics["perplexity"] = perplexity + + trainer.log_metrics("eval", metrics) + trainer.save_metrics("eval", metrics) + + kwargs = { + "finetuned_from": model_args.model_name_or_path, + "tasks": "text-generation", + } + if data_args.dataset_name is not None: + kwargs["dataset_tags"] = data_args.dataset_name + if data_args.dataset_config_name is not None: + kwargs["dataset_args"] = data_args.dataset_config_name + kwargs[ + "dataset" + ] = f"{data_args.dataset_name} {data_args.dataset_config_name}" + else: + kwargs["dataset"] = data_args.dataset_name + + if training_args.push_to_hub: + trainer.push_to_hub(**kwargs) + else: + trainer.create_model_card(**kwargs) + + +def _mp_fn(index): + # For xla_spawn (TPUs) + main() + + +if __name__ == "__main__": + main() diff --git a/demo-notebooks/guided-demos/mnist.py b/demo-notebooks/guided-demos/mnist.py new file mode 100644 index 000000000..6eb663dc7 --- /dev/null +++ b/demo-notebooks/guided-demos/mnist.py @@ -0,0 +1,160 @@ +# Copyright 2022 IBM, Red Hat +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +# In[] +import os + +import torch +from pytorch_lightning import LightningModule, Trainer +from pytorch_lightning.callbacks.progress import TQDMProgressBar +from pytorch_lightning.loggers import CSVLogger +from torch import nn +from torch.nn import functional as F +from torch.utils.data import DataLoader, random_split +from torchmetrics import Accuracy +from torchvision import transforms +from torchvision.datasets import MNIST + +PATH_DATASETS = os.environ.get("PATH_DATASETS", ".") +BATCH_SIZE = 256 if torch.cuda.is_available() else 64 +# %% + +print("prior to running the trainer") +print("MASTER_ADDR: is ", os.getenv("MASTER_ADDR")) +print("MASTER_PORT: is ", os.getenv("MASTER_PORT")) + + +class LitMNIST(LightningModule): + def __init__(self, data_dir=PATH_DATASETS, hidden_size=64, learning_rate=2e-4): + super().__init__() + + # Set our init args as class attributes + self.data_dir = data_dir + self.hidden_size = hidden_size + self.learning_rate = learning_rate + + # Hardcode some dataset specific attributes + self.num_classes = 10 + self.dims = (1, 28, 28) + channels, width, height = self.dims + self.transform = transforms.Compose( + [ + transforms.ToTensor(), + transforms.Normalize((0.1307,), (0.3081,)), + ] + ) + + # Define PyTorch model + self.model = nn.Sequential( + nn.Flatten(), + nn.Linear(channels * width * height, hidden_size), + nn.ReLU(), + nn.Dropout(0.1), + nn.Linear(hidden_size, hidden_size), + nn.ReLU(), + nn.Dropout(0.1), + nn.Linear(hidden_size, self.num_classes), + ) + + self.val_accuracy = Accuracy() + self.test_accuracy = Accuracy() + + def forward(self, x): + x = self.model(x) + return F.log_softmax(x, dim=1) + + def training_step(self, batch, batch_idx): + x, y = batch + logits = self(x) + loss = F.nll_loss(logits, y) + return loss + + def validation_step(self, batch, batch_idx): + x, y = batch + logits = self(x) + loss = F.nll_loss(logits, y) + preds = torch.argmax(logits, dim=1) + self.val_accuracy.update(preds, y) + + # Calling self.log will surface up scalars for you in TensorBoard + self.log("val_loss", loss, prog_bar=True) + self.log("val_acc", self.val_accuracy, prog_bar=True) + + def test_step(self, batch, batch_idx): + x, y = batch + logits = self(x) + loss = F.nll_loss(logits, y) + preds = torch.argmax(logits, dim=1) + self.test_accuracy.update(preds, y) + + # Calling self.log will surface up scalars for you in TensorBoard + self.log("test_loss", loss, prog_bar=True) + self.log("test_acc", self.test_accuracy, prog_bar=True) + + def configure_optimizers(self): + optimizer = torch.optim.Adam(self.parameters(), lr=self.learning_rate) + return optimizer + + #################### + # DATA RELATED HOOKS + #################### + + def prepare_data(self): + # download + print("Downloading MNIST dataset...") + MNIST(self.data_dir, train=True, download=True) + MNIST(self.data_dir, train=False, download=True) + + def setup(self, stage=None): + # Assign train/val datasets for use in dataloaders + if stage == "fit" or stage is None: + mnist_full = MNIST(self.data_dir, train=True, transform=self.transform) + self.mnist_train, self.mnist_val = random_split(mnist_full, [55000, 5000]) + + # Assign test dataset for use in dataloader(s) + if stage == "test" or stage is None: + self.mnist_test = MNIST( + self.data_dir, train=False, transform=self.transform + ) + + def train_dataloader(self): + return DataLoader(self.mnist_train, batch_size=BATCH_SIZE) + + def val_dataloader(self): + return DataLoader(self.mnist_val, batch_size=BATCH_SIZE) + + def test_dataloader(self): + return DataLoader(self.mnist_test, batch_size=BATCH_SIZE) + + +# Init DataLoader from MNIST Dataset + +model = LitMNIST() + +print("GROUP: ", int(os.environ.get("GROUP_WORLD_SIZE", 1))) +print("LOCAL: ", int(os.environ.get("LOCAL_WORLD_SIZE", 1))) + +# Initialize a trainer +trainer = Trainer( + accelerator="auto", + # devices=1 if torch.cuda.is_available() else None, # limiting got iPython runs + max_epochs=5, + callbacks=[TQDMProgressBar(refresh_rate=20)], + num_nodes=int(os.environ.get("GROUP_WORLD_SIZE", 1)), + devices=int(os.environ.get("LOCAL_WORLD_SIZE", 1)), + strategy="ddp", +) + +# Train the model ⚡ +trainer.fit(model) diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/0_basic_ray.ipynb b/demo-notebooks/guided-demos/notebook-ex-outputs/0_basic_ray.ipynb new file mode 100644 index 000000000..0199b1fb7 --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/0_basic_ray.ipynb @@ -0,0 +1,366 @@ +{ + "cells": [ + { + "attachments": {}, + "cell_type": "markdown", + "id": "8d4a42f6", + "metadata": {}, + "source": [ + "In this first notebook, we will go through the basics of using the SDK to:\n", + " - Spin up a Ray cluster with our desired resources\n", + " - View the status and specs of our Ray cluster\n", + " - Take down the Ray cluster when finished" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "b55bc3ea-4ce3-49bf-bb1f-e209de8ca47a", + "metadata": {}, + "outputs": [], + "source": [ + "# Import pieces from codeflare-sdk\n", + "from codeflare_sdk.cluster.cluster import Cluster, ClusterConfiguration\n", + "from codeflare_sdk.cluster.auth import TokenAuthentication" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "614daa0c", + "metadata": {}, + "outputs": [], + "source": [ + "# Create authentication object for oc user permissions\n", + "auth = TokenAuthentication(\n", + " token = \"XXXXX\",\n", + " server = \"XXXXX\",\n", + " skip_tls=False\n", + ")\n", + "auth.login()" + ] + }, + { + "cell_type": "markdown", + "id": "bc27f84c", + "metadata": {}, + "source": [ + "Here, we want to define our cluster by specifying the resources we require for our batch workload. Below, we define our cluster object (which generates a corresponding AppWrapper)." + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "0f4bc870-091f-4e11-9642-cba145710159", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Written to: raytest.yaml\n" + ] + } + ], + "source": [ + "# Create and configure our cluster object (and appwrapper)\n", + "cluster = Cluster(ClusterConfiguration(\n", + " name='raytest',\n", + " namespace='default',\n", + " min_worker=2,\n", + " max_worker=2,\n", + " min_cpus=1,\n", + " max_cpus=1,\n", + " min_memory=4,\n", + " max_memory=4,\n", + " gpu=0,\n", + " instascale=False\n", + "))" + ] + }, + { + "cell_type": "markdown", + "id": "12eef53c", + "metadata": {}, + "source": [ + "Next, we want to bring our cluster up, so we call the `up()` function below to submit our cluster AppWrapper yaml onto the MCAD queue, and begin the process of obtaining our resource cluster." + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "f0884bbc-c224-4ca0-98a0-02dfa09c2200", + "metadata": {}, + "outputs": [], + "source": [ + "# Bring up the cluster\n", + "cluster.up()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "657ebdfb", + "metadata": {}, + "source": [ + "Now, we want to check on the status of our resource cluster, and wait until it is finally ready for use." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "3c1b4311-2e61-44c9-8225-87c2db11363d", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
╭───────────────────────╮\n",
+       "│    🚀 Cluster Queue   │\n",
+       "│       Status 🚀       │\n",
+       "│ +---------+---------+ │\n",
+       "│ | Name    | Status  | │\n",
+       "│ +=========+=========+ │\n",
+       "│ | raytest | pending | │\n",
+       "│ |         |         | │\n",
+       "│ +---------+---------+ │\n",
+       "╰───────────────────────╯\n",
+       "
\n" + ], + "text/plain": [ + "╭───────────────────────╮\n", + "│ \u001b[3m \u001b[0m\u001b[1;3m 🚀 Cluster Queue\u001b[0m\u001b[3m \u001b[0m │\n", + "│ \u001b[3m \u001b[0m\u001b[1;3mStatus 🚀\u001b[0m\u001b[3m \u001b[0m │\n", + "│ +---------+---------+ │\n", + "│ |\u001b[1m \u001b[0m\u001b[1mName \u001b[0m\u001b[1m \u001b[0m|\u001b[1m \u001b[0m\u001b[1mStatus \u001b[0m\u001b[1m \u001b[0m| │\n", + "│ +=========+=========+ │\n", + "│ |\u001b[36m \u001b[0m\u001b[36mraytest\u001b[0m\u001b[36m \u001b[0m|\u001b[35m \u001b[0m\u001b[35mpending\u001b[0m\u001b[35m \u001b[0m| │\n", + "│ |\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m|\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m| │\n", + "│ +---------+---------+ │\n", + "╰───────────────────────╯\n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "(, False)" + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cluster.status()" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "a99d5aff", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Waiting for requested resources to be set up...\n", + "Requested cluster up and running!\n" + ] + } + ], + "source": [ + "cluster.wait_ready()" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "df71c1ed", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
                  🚀 CodeFlare Cluster Status 🚀                  \n",
+       "                                                                  \n",
+       " ╭──────────────────────────────────────────────────────────────╮ \n",
+       " │   Name                                                       │ \n",
+       " │   raytest                                        Active ✅   │ \n",
+       " │                                                              │ \n",
+       " │   URI: ray://raytest-head-svc.default.svc:10001              │ \n",
+       " │                                                              │ \n",
+       " │   Dashboard🔗                                                │ \n",
+       " │                                                              │ \n",
+       " ╰──────────────────────────────────────────────────────────────╯ \n",
+       "
\n" + ], + "text/plain": [ + "\u001b[3m \u001b[0m\u001b[1;3m 🚀 CodeFlare Cluster Status 🚀\u001b[0m\u001b[3m \u001b[0m\n", + "\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\n", + " ╭──────────────────────────────────────────────────────────────╮ \n", + " │ \u001b[1;37;42mName\u001b[0m │ \n", + " │ \u001b[1;4mraytest\u001b[0m Active ✅ │ \n", + " │ │ \n", + " │ \u001b[1mURI:\u001b[0m ray://raytest-head-svc.default.svc:10001 │ \n", + " │ │ \n", + " │ \u001b]8;id=786870;ray-dashboard-raytest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org\u001b\\\u001b[4;34mDashboard🔗\u001b[0m\u001b]8;;\u001b\\ │ \n", + " │ │ \n", + " ╰──────────────────────────────────────────────────────────────╯ \n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "(, True)" + ] + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cluster.status()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "b3a55fe4", + "metadata": {}, + "source": [ + "Let's quickly verify that the specs of the cluster are as expected." + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "7fd45bc5-03c0-4ae5-9ec5-dd1c30f1a084", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
                  🚀 CodeFlare Cluster Details 🚀                 \n",
+       "                                                                  \n",
+       " ╭──────────────────────────────────────────────────────────────╮ \n",
+       " │   Name                                                       │ \n",
+       " │   raytest                                        Active ✅   │ \n",
+       " │                                                              │ \n",
+       " │   URI: ray://raytest-head-svc.default.svc:10001              │ \n",
+       " │                                                              │ \n",
+       " │   Dashboard🔗                                                │ \n",
+       " │                                                              │ \n",
+       " │                      Cluster Resources                       │ \n",
+       " │   ╭─ Workers ──╮  ╭───────── Worker specs(each) ─────────╮   │ \n",
+       " │   │  Min  Max  │  │  Memory      CPU         GPU         │   │ \n",
+       " │   │            │  │                                      │   │ \n",
+       " │   │  2    2    │  │  4~4         1           0           │   │ \n",
+       " │   │            │  │                                      │   │ \n",
+       " │   ╰────────────╯  ╰──────────────────────────────────────╯   │ \n",
+       " ╰──────────────────────────────────────────────────────────────╯ \n",
+       "
\n" + ], + "text/plain": [ + "\u001b[3m \u001b[0m\u001b[1;3m 🚀 CodeFlare Cluster Details 🚀\u001b[0m\u001b[3m \u001b[0m\n", + "\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\n", + " ╭──────────────────────────────────────────────────────────────╮ \n", + " │ \u001b[1;37;42mName\u001b[0m │ \n", + " │ \u001b[1;4mraytest\u001b[0m Active ✅ │ \n", + " │ │ \n", + " │ \u001b[1mURI:\u001b[0m ray://raytest-head-svc.default.svc:10001 │ \n", + " │ │ \n", + " │ \u001b]8;id=611457;http://ray-dashboard-raytest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org\u001b\\\u001b[4;34mDashboard🔗\u001b[0m\u001b]8;;\u001b\\ │ \n", + " │ │ \n", + " │ \u001b[3m Cluster Resources \u001b[0m │ \n", + " │ ╭─ Workers ──╮ ╭───────── Worker specs(each) ─────────╮ │ \n", + " │ │ \u001b[1m \u001b[0m\u001b[1mMin\u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mMax\u001b[0m\u001b[1m \u001b[0m │ │ \u001b[1m \u001b[0m\u001b[1mMemory \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mCPU \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mGPU \u001b[0m\u001b[1m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m2 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m2 \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m4~4 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m1 \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m0 \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ ╰────────────╯ ╰──────────────────────────────────────╯ │ \n", + " ╰──────────────────────────────────────────────────────────────╯ \n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "RayCluster(name='raytest', status=, min_workers=2, max_workers=2, worker_mem_min=4, worker_mem_max=4, worker_cpu=1, worker_gpu=0, namespace='default', dashboard='http://ray-dashboard-raytest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org')" + ] + }, + "execution_count": 8, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cluster.details()" + ] + }, + { + "cell_type": "markdown", + "id": "5af8cd32", + "metadata": {}, + "source": [ + "Finally, we bring our resource cluster down and release/terminate the associated resources, bringing everything back to the way it was before our cluster was brought up." + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "5f36db0f-31f6-4373-9503-dc3c1c4c3f57", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.down()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0d41b90e", + "metadata": {}, + "outputs": [], + "source": [ + "auth.logout()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "sdktest", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.13" + }, + "vscode": { + "interpreter": { + "hash": "ae72ec8d55aeb4773d9bab14ab14ec6c410f2dd8be83850b7c2732f479ead773" + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/1_basic_instascale.ipynb b/demo-notebooks/guided-demos/notebook-ex-outputs/1_basic_instascale.ipynb new file mode 100644 index 000000000..ed5716d5c --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/1_basic_instascale.ipynb @@ -0,0 +1,253 @@ +{ + "cells": [ + { + "attachments": {}, + "cell_type": "markdown", + "id": "9865ee8c", + "metadata": {}, + "source": [ + "In this second notebook, we will go over the basics of using InstaScale to scale up/down necessary resources that are not currently available on your OpenShift Cluster (in cloud environments)." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "b55bc3ea-4ce3-49bf-bb1f-e209de8ca47a", + "metadata": {}, + "outputs": [], + "source": [ + "# Import pieces from codeflare-sdk\n", + "from codeflare_sdk.cluster.cluster import Cluster, ClusterConfiguration\n", + "from codeflare_sdk.cluster.auth import TokenAuthentication" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "614daa0c", + "metadata": {}, + "outputs": [], + "source": [ + "# Create authentication object for oc user permissions\n", + "auth = TokenAuthentication(\n", + " token = \"XXXXX\",\n", + " server = \"XXXXX\",\n", + " skip_tls=False\n", + ")\n", + "auth.login()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "bc27f84c", + "metadata": {}, + "source": [ + "This time, we are working in a cloud environment, and our OpenShift cluster does not have the resources needed for our desired workloads. We will use InstaScale to dynamically scale-up guaranteed resources based on our request (that will also automatically scale-down when we are finished working):" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "0f4bc870-091f-4e11-9642-cba145710159", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Written to: instascaletest.yaml\n" + ] + } + ], + "source": [ + "# Create and configure our cluster object (and appwrapper)\n", + "cluster = Cluster(ClusterConfiguration(\n", + " name='instascaletest',\n", + " namespace='default',\n", + " min_worker=2,\n", + " max_worker=2,\n", + " min_cpus=2,\n", + " max_cpus=2,\n", + " min_memory=8,\n", + " max_memory=8,\n", + " gpu=1,\n", + " instascale=True, # InstaScale now enabled, will scale OCP cluster to guarantee resource request\n", + " machine_types=[\"m5.xlarge\", \"g4dn.xlarge\"] # Head, worker AWS machine types desired\n", + "))" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "12eef53c", + "metadata": {}, + "source": [ + "Same as last time, we will bring the cluster up, wait for it to be ready, and confirm that the specs are as-requested:" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "f0884bbc-c224-4ca0-98a0-02dfa09c2200", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Waiting for requested resources to be set up...\n", + "Requested cluster up and running!\n" + ] + } + ], + "source": [ + "# Bring up the cluster\n", + "cluster.up()\n", + "cluster.wait_ready()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "6abfe904", + "metadata": {}, + "source": [ + "While the resources are being scaled, we can also go into the console and take a look at the InstaScale logs, as well as the new machines/nodes spinning up.\n", + "\n", + "Once the cluster is ready, we can confirm the specs:" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "7fd45bc5-03c0-4ae5-9ec5-dd1c30f1a084", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
                     🚀 CodeFlare Cluster Details 🚀                     \n",
+       "                                                                         \n",
+       " ╭─────────────────────────────────────────────────────────────────────╮ \n",
+       " │   Name                                                              │ \n",
+       " │   instascaletest                                        Active ✅   │ \n",
+       " │                                                                     │ \n",
+       " │   URI: ray://instascaletest-head-svc.default.svc:10001              │ \n",
+       " │                                                                     │ \n",
+       " │   Dashboard🔗                                                       │ \n",
+       " │                                                                     │ \n",
+       " │                      Cluster Resources                              │ \n",
+       " │   ╭─ Workers ──╮  ╭───────── Worker specs(each) ─────────╮          │ \n",
+       " │   │  Min  Max  │  │  Memory      CPU         GPU         │          │ \n",
+       " │   │            │  │                                      │          │ \n",
+       " │   │  2    2    │  │  8~8         2           1           │          │ \n",
+       " │   │            │  │                                      │          │ \n",
+       " │   ╰────────────╯  ╰──────────────────────────────────────╯          │ \n",
+       " ╰─────────────────────────────────────────────────────────────────────╯ \n",
+       "
\n" + ], + "text/plain": [ + "\u001b[3m \u001b[0m\u001b[1;3m 🚀 CodeFlare Cluster Details 🚀\u001b[0m\u001b[3m \u001b[0m\n", + "\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\n", + " ╭─────────────────────────────────────────────────────────────────────╮ \n", + " │ \u001b[1;37;42mName\u001b[0m │ \n", + " │ \u001b[1;4minstascaletest\u001b[0m Active ✅ │ \n", + " │ │ \n", + " │ \u001b[1mURI:\u001b[0m ray://instascaletest-head-svc.default.svc:10001 │ \n", + " │ │ \n", + " │ \u001b]8;id=164702;http://ray-dashboard-instascaletest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org\u001b\\\u001b[4;34mDashboard🔗\u001b[0m\u001b]8;;\u001b\\ │ \n", + " │ │ \n", + " │ \u001b[3m Cluster Resources \u001b[0m │ \n", + " │ ╭─ Workers ──╮ ╭───────── Worker specs(each) ─────────╮ │ \n", + " │ │ \u001b[1m \u001b[0m\u001b[1mMin\u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mMax\u001b[0m\u001b[1m \u001b[0m │ │ \u001b[1m \u001b[0m\u001b[1mMemory \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mCPU \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mGPU \u001b[0m\u001b[1m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m2 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m2 \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m8~8 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m2 \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m1 \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ ╰────────────╯ ╰──────────────────────────────────────╯ │ \n", + " ╰─────────────────────────────────────────────────────────────────────╯ \n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "RayCluster(name='instascaletest', status=, min_workers=2, max_workers=2, worker_mem_min=8, worker_mem_max=8, worker_cpu=2, worker_gpu=1, namespace='default', dashboard='http://ray-dashboard-instascaletest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org')" + ] + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cluster.details()" + ] + }, + { + "cell_type": "markdown", + "id": "5af8cd32", + "metadata": {}, + "source": [ + "Finally, we bring our resource cluster down and release/terminate the associated resources, bringing everything back to the way it was before our cluster was brought up." + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "5f36db0f-31f6-4373-9503-dc3c1c4c3f57", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.down()" + ] + }, + { + "attachments": {}, + "cell_type": "markdown", + "id": "c883caea", + "metadata": {}, + "source": [ + "Once again, we can look at the machines/nodes and see that everything has been successfully scaled down!" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "0d41b90e", + "metadata": {}, + "outputs": [], + "source": [ + "auth.logout()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.9.13" + }, + "vscode": { + "interpreter": { + "hash": "f9f85f796d01129d0dd105a088854619f454435301f6ffec2fea96ecbd9be4ac" + } + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/2_basic_jobs.ipynb b/demo-notebooks/guided-demos/notebook-ex-outputs/2_basic_jobs.ipynb new file mode 100644 index 000000000..045b69b3c --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/2_basic_jobs.ipynb @@ -0,0 +1,454 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "464af595", + "metadata": {}, + "source": [ + "In this third notebook, we will go over the basics of submitting jobs via the SDK, either to a Ray cluster or directly to MCAD." + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "b55bc3ea-4ce3-49bf-bb1f-e209de8ca47a", + "metadata": {}, + "outputs": [], + "source": [ + "# Import pieces from codeflare-sdk\n", + "from codeflare_sdk.cluster.cluster import Cluster, ClusterConfiguration\n", + "from codeflare_sdk.cluster.auth import TokenAuthentication" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "614daa0c", + "metadata": {}, + "outputs": [], + "source": [ + "# Create authentication object for oc user permissions\n", + "auth = TokenAuthentication(\n", + " token = \"XXXXX\",\n", + " server = \"XXXXX\",\n", + " skip_tls=False\n", + ")\n", + "auth.login()" + ] + }, + { + "cell_type": "markdown", + "id": "bc27f84c", + "metadata": {}, + "source": [ + "Let's start by running through the same cluster setup as before:" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "0f4bc870-091f-4e11-9642-cba145710159", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Written to: jobtest.yaml\n" + ] + } + ], + "source": [ + "# Create and configure our cluster object (and appwrapper)\n", + "cluster = Cluster(ClusterConfiguration(\n", + " name='jobtest',\n", + " namespace='default',\n", + " min_worker=2,\n", + " max_worker=2,\n", + " min_cpus=1,\n", + " max_cpus=1,\n", + " min_memory=4,\n", + " max_memory=4,\n", + " gpu=0,\n", + " instascale=False\n", + "))" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "f0884bbc-c224-4ca0-98a0-02dfa09c2200", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Waiting for requested resources to be set up...\n", + "Requested cluster up and running!\n" + ] + } + ], + "source": [ + "# Bring up the cluster\n", + "cluster.up()\n", + "cluster.wait_ready()" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "df71c1ed", + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
                  🚀 CodeFlare Cluster Details 🚀                 \n",
+       "                                                                  \n",
+       " ╭──────────────────────────────────────────────────────────────╮ \n",
+       " │   Name                                                       │ \n",
+       " │   jobtest                                        Active ✅   │ \n",
+       " │                                                              │ \n",
+       " │   URI: ray://jobtest-head-svc.default.svc:10001              │ \n",
+       " │                                                              │ \n",
+       " │   Dashboard🔗                                                │ \n",
+       " │                                                              │ \n",
+       " │                      Cluster Resources                       │ \n",
+       " │   ╭─ Workers ──╮  ╭───────── Worker specs(each) ─────────╮   │ \n",
+       " │   │  Min  Max  │  │  Memory      CPU         GPU         │   │ \n",
+       " │   │            │  │                                      │   │ \n",
+       " │   │  2    2    │  │  4~4         1           0           │   │ \n",
+       " │   │            │  │                                      │   │ \n",
+       " │   ╰────────────╯  ╰──────────────────────────────────────╯   │ \n",
+       " ╰──────────────────────────────────────────────────────────────╯ \n",
+       "
\n" + ], + "text/plain": [ + "\u001b[3m \u001b[0m\u001b[1;3m 🚀 CodeFlare Cluster Details 🚀\u001b[0m\u001b[3m \u001b[0m\n", + "\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\n", + " ╭──────────────────────────────────────────────────────────────╮ \n", + " │ \u001b[1;37;42mName\u001b[0m │ \n", + " │ \u001b[1;4mjobtest\u001b[0m Active ✅ │ \n", + " │ │ \n", + " │ \u001b[1mURI:\u001b[0m ray://jobtest-head-svc.default.svc:10001 │ \n", + " │ │ \n", + " │ \u001b]8;id=793109;http://ray-dashboard-jobtest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org\u001b\\\u001b[4;34mDashboard🔗\u001b[0m\u001b]8;;\u001b\\ │ \n", + " │ │ \n", + " │ \u001b[3m Cluster Resources \u001b[0m │ \n", + " │ ╭─ Workers ──╮ ╭───────── Worker specs(each) ─────────╮ │ \n", + " │ │ \u001b[1m \u001b[0m\u001b[1mMin\u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mMax\u001b[0m\u001b[1m \u001b[0m │ │ \u001b[1m \u001b[0m\u001b[1mMemory \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mCPU \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mGPU \u001b[0m\u001b[1m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m2 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m2 \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m4~4 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m1 \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m0 \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ ╰────────────╯ ╰──────────────────────────────────────╯ │ \n", + " ╰──────────────────────────────────────────────────────────────╯ \n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "RayCluster(name='jobtest', status=, min_workers=2, max_workers=2, worker_mem_min=4, worker_mem_max=4, worker_cpu=1, worker_gpu=0, namespace='default', dashboard='http://ray-dashboard-jobtest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org')" + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cluster.details()" + ] + }, + { + "cell_type": "markdown", + "id": "33663f47", + "metadata": {}, + "source": [ + "This time, however, we are going to use the CodeFlare SDK to submit batch jobs via TorchX, either to the Ray cluster we have just brought up, or directly to MCAD." + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "c7b4f232", + "metadata": {}, + "outputs": [], + "source": [ + "from codeflare_sdk.job.jobs import DDPJobDefinition" + ] + }, + { + "cell_type": "markdown", + "id": "83d77b74", + "metadata": {}, + "source": [ + "First, let's begin by submitting to Ray, training a basic NN on the MNIST dataset:" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "8c2c5138", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "The Ray scheduler does not support port mapping.\n" + ] + } + ], + "source": [ + "jobdef = DDPJobDefinition(\n", + " name=\"mnisttest\",\n", + " script=\"mnist.py\",\n", + " scheduler_args={\"requirements\": \"requirements.txt\"}\n", + ")\n", + "job = jobdef.submit(cluster)" + ] + }, + { + "cell_type": "markdown", + "id": "5b9ae53a", + "metadata": {}, + "source": [ + "Now we can take a look at the status of our submitted job, as well as the logs:" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "6e36c3d9", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "AppStatus:\n", + " msg: !!python/object/apply:ray.dashboard.modules.job.common.JobStatus\n", + " - SUCCEEDED\n", + " num_restarts: -1\n", + " roles:\n", + " - replicas:\n", + " - hostname: \n", + " id: 0\n", + " role: ray\n", + " state: !!python/object/apply:torchx.specs.api.AppState\n", + " - 4\n", + " structured_error_msg: \n", + " role: ray\n", + " state: SUCCEEDED (4)\n", + " structured_error_msg: \n", + " ui_url: null" + ] + }, + "execution_count": 10, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "job.status()" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "834cfb5c", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'[RayActor(name=\\'mnist\\', command=[\\'bash\\', \\'-c\\', \"torchrun --rdzv_backend static --rdzv_endpoint $TORCHX_RANK0_HOST:49782 --rdzv_id \\'mnisttest-f50nbxs5pnwwlc\\' --nnodes 2 --nproc_per_node 1 --node_rank \\'0\\' --tee 3 --role \\'\\' mnist.py\"], env={\\'TORCHX_TRACKING_EXPERIMENT_NAME\\': \\'default-experiment\\', \\'LOGLEVEL\\': \\'WARNING\\', \\'TORCHX_JOB_ID\\': \\'ray://torchx/mnisttest-f50nbxs5pnwwlc\\'}, num_cpus=1, num_gpus=0, min_replicas=2), RayActor(name=\\'mnist\\', command=[\\'bash\\', \\'-c\\', \"torchrun --rdzv_backend static --rdzv_endpoint $TORCHX_RANK0_HOST:49782 --rdzv_id \\'mnisttest-f50nbxs5pnwwlc\\' --nnodes 2 --nproc_per_node 1 --node_rank \\'1\\' --tee 3 --role \\'\\' mnist.py\"], env={\\'TORCHX_TRACKING_EXPERIMENT_NAME\\': \\'default-experiment\\', \\'LOGLEVEL\\': \\'WARNING\\', \\'TORCHX_JOB_ID\\': \\'ray://torchx/mnisttest-f50nbxs5pnwwlc\\'}, num_cpus=1, num_gpus=0, min_replicas=2)]\\n2023-04-26 11:30:03,092\\tINFO worker.py:1230 -- Using address 10.129.0.182:6379 set in the environment variable RAY_ADDRESS\\n2023-04-26 11:30:03,092\\tINFO worker.py:1342 -- Connecting to existing Ray cluster at address: 10.129.0.182:6379...\\n2023-04-26 11:30:03,097\\tINFO worker.py:1519 -- Connected to Ray cluster. View the dashboard at \\x1b[1m\\x1b[32mhttp://10.129.0.182:8265 \\x1b[39m\\x1b[22m\\nWaiting for minimum placement group to start.\\nSuccessfully created placement groups\\nrdzv_endpoint set to 10.129.0.182 for actor e0a0863650338de1363ea91502000000\\nrdzv_endpoint set to 10.129.0.182 for actor 5d5c64cf765c7982d83e6b3202000000\\nSuccessfully placed command actors\\nEntering main loop, start executing the script on worker nodes\\nrunning ray.wait on [ObjectRef(e082c90ab8422b00e0a0863650338de1363ea9150200000001000000), ObjectRef(ce868e48e2fa9a945d5c64cf765c7982d83e6b320200000001000000)]\\nrunning ray.wait on [ObjectRef(ce868e48e2fa9a945d5c64cf765c7982d83e6b320200000001000000), ObjectRef(f81ec6ff838b16dbe0a0863650338de1363ea9150200000001000000)]\\nrunning ray.wait on [ObjectRef(f81ec6ff838b16dbe0a0863650338de1363ea9150200000001000000), ObjectRef(32b0eec39cfa87ac5d5c64cf765c7982d83e6b320200000001000000)]\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:prior to running the trainer\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:MASTER_ADDR: is 10.129.0.182\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:MASTER_PORT: is 49782\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:GROUP: 2\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:LOCAL: 1\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Downloading MNIST dataset...\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Downloading http://yann.lecun.com/exdb/mnist/train-images-idx3-ubyte.gz to ./MNIST/raw/train-images-idx3-ubyte.gz\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Extracting ./MNIST/raw/train-images-idx3-ubyte.gz to ./MNIST/raw\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Downloading http://yann.lecun.com/exdb/mnist/train-labels-idx1-ubyte.gz to ./MNIST/raw/train-labels-idx1-ubyte.gz\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Extracting ./MNIST/raw/train-labels-idx1-ubyte.gz to ./MNIST/raw\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Downloading http://yann.lecun.com/exdb/mnist/t10k-images-idx3-ubyte.gz to ./MNIST/raw/t10k-images-idx3-ubyte.gz\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Extracting ./MNIST/raw/t10k-images-idx3-ubyte.gz to ./MNIST/raw\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Downloading http://yann.lecun.com/exdb/mnist/t10k-labels-idx1-ubyte.gz to ./MNIST/raw/t10k-labels-idx1-ubyte.gz\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Extracting ./MNIST/raw/t10k-labels-idx1-ubyte.gz to ./MNIST/raw\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Validation sanity check: 0it [00:00, ?it/s]\\n\\x1b[2m\\x1b[36m(CommandActor pid=327)\\x1b[0m [0]:Validation sanity check: 0%| | 0/2 [00:00\n", + " num_restarts: -1\n", + " roles:\n", + " - replicas:\n", + " - hostname: ''\n", + " id: 0\n", + " role: mnist\n", + " state: !!python/object/apply:torchx.specs.api.AppState\n", + " - 3\n", + " structured_error_msg: \n", + " role: mnist\n", + " state: RUNNING (3)\n", + " structured_error_msg: \n", + " ui_url: null" + ] + }, + "execution_count": 18, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "job.status()" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "36d7ea97", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'2023-04-26T18:33:22.956866570Z [0]:GPU available: False, used: False\\n2023-04-26T18:33:22.956866570Z [0]:TPU available: False, using: 0 TPU cores\\n2023-04-26T18:33:22.956866570Z [0]:IPU available: False, using: 0 IPUs\\n2023-04-26T18:33:23.057069471Z [0]:\\n2023-04-26T18:33:23.157240736Z [0]: 0%| | 0/9912422 [00:00 🚀 CodeFlare Cluster Details 🚀 \n", + " \n", + " ╭──────────────────────────────────────────────────────────────────────╮ \n", + " │ Name │ \n", + " │ interactivetest Active ✅ │ \n", + " │ │ \n", + " │ URI: ray://interactivetest-head-svc.default.svc:10001 │ \n", + " │ │ \n", + " │ Dashboard🔗 │ \n", + " │ │ \n", + " │ Cluster Resources │ \n", + " │ ╭─ Workers ──╮ ╭───────── Worker specs(each) ─────────╮ │ \n", + " │ │ Min Max │ │ Memory CPU GPU │ │ \n", + " │ │ │ │ │ │ \n", + " │ │ 2 2 │ │ 8~8 2 1 │ │ \n", + " │ │ │ │ │ │ \n", + " │ ╰────────────╯ ╰──────────────────────────────────────╯ │ \n", + " ╰──────────────────────────────────────────────────────────────────────╯ \n", + "\n" + ], + "text/plain": [ + "\u001b[3m \u001b[0m\u001b[1;3m 🚀 CodeFlare Cluster Details 🚀\u001b[0m\u001b[3m \u001b[0m\n", + "\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\n", + " ╭──────────────────────────────────────────────────────────────────────╮ \n", + " │ \u001b[1;37;42mName\u001b[0m │ \n", + " │ \u001b[1;4minteractivetest\u001b[0m Active ✅ │ \n", + " │ │ \n", + " │ \u001b[1mURI:\u001b[0m ray://interactivetest-head-svc.default.svc:10001 │ \n", + " │ │ \n", + " │ \u001b]8;id=789787;http://ray-dashboard-interactivetest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org\u001b\\\u001b[4;34mDashboard🔗\u001b[0m\u001b]8;;\u001b\\ │ \n", + " │ │ \n", + " │ \u001b[3m Cluster Resources \u001b[0m │ \n", + " │ ╭─ Workers ──╮ ╭───────── Worker specs(each) ─────────╮ │ \n", + " │ │ \u001b[1m \u001b[0m\u001b[1mMin\u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mMax\u001b[0m\u001b[1m \u001b[0m │ │ \u001b[1m \u001b[0m\u001b[1mMemory \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mCPU \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mGPU \u001b[0m\u001b[1m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m2 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m2 \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m8~8 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m2 \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m1 \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ ╰────────────╯ ╰──────────────────────────────────────╯ │ \n", + " ╰──────────────────────────────────────────────────────────────────────╯ \n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "RayCluster(name='interactivetest', status=, min_workers=2, max_workers=2, worker_mem_min=8, worker_mem_max=8, worker_cpu=2, worker_gpu=1, namespace='default', dashboard='http://ray-dashboard-interactivetest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org')" + ] + }, + "execution_count": 5, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cluster.details()" + ] + }, + { + "cell_type": "markdown", + "id": "33663f47", + "metadata": {}, + "source": [ + "This time we will demonstrate another potential method of use: working with the Ray cluster interactively.\n", + "\n", + "Using the SDK, we can get both the Ray cluster URI and dashboard URI:" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "c1719bca", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "http://ray-dashboard-interactivetest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org\n", + "ray://interactivetest-head-svc.default.svc:10001\n" + ] + } + ], + "source": [ + "ray_dashboard_uri = cluster.cluster_dashboard_uri()\n", + "ray_cluster_uri = cluster.cluster_uri()\n", + "print(ray_dashboard_uri)\n", + "print(ray_cluster_uri)" + ] + }, + { + "cell_type": "markdown", + "id": "2a2aca6a", + "metadata": {}, + "source": [ + "Now we can connect directly to our Ray cluster via the Ray python client:" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "300146dc", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Ray cluster is up and running: True\n" + ] + } + ], + "source": [ + "#before proceeding make sure the cluster exists and the uri is not empty\n", + "assert ray_cluster_uri, \"Ray cluster needs to be started and set before proceeding\"\n", + "\n", + "import ray\n", + "from ray.air.config import ScalingConfig\n", + "\n", + "# reset the ray context in case there's already one. \n", + "ray.shutdown()\n", + "# establish connection to ray cluster\n", + "\n", + "#install additionall libraries that will be required for model training\n", + "runtime_env = {\"pip\": [\"transformers\", \"datasets\", \"evaluate\", \"pyarrow<7.0.0\"]}\n", + "\n", + "ray.init(address=f'{ray_cluster_uri}', runtime_env=runtime_env)\n", + "\n", + "print(\"Ray cluster is up and running: \", ray.is_initialized())" + ] + }, + { + "cell_type": "markdown", + "id": "9711030b", + "metadata": {}, + "source": [ + "Now that we are connected (and have passed in some package requirements), let's try writing some training code for a DistilBERT transformer model via HuggingFace (using IMDB dataset):" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "1b36e0d9", + "metadata": {}, + "outputs": [], + "source": [ + "@ray.remote\n", + "def train_fn():\n", + " from datasets import load_dataset\n", + " import transformers\n", + " from transformers import AutoTokenizer, TrainingArguments\n", + " from transformers import AutoModelForSequenceClassification\n", + " import numpy as np\n", + " from datasets import load_metric\n", + " import ray\n", + " from ray import tune\n", + " from ray.train.huggingface import HuggingFaceTrainer\n", + "\n", + " dataset = load_dataset(\"imdb\")\n", + " tokenizer = AutoTokenizer.from_pretrained(\"distilbert-base-uncased\")\n", + "\n", + " def tokenize_function(examples):\n", + " return tokenizer(examples[\"text\"], padding=\"max_length\", truncation=True)\n", + "\n", + " tokenized_datasets = dataset.map(tokenize_function, batched=True)\n", + "\n", + " #using a fraction of dataset but you can run with the full dataset\n", + " small_train_dataset = tokenized_datasets[\"train\"].shuffle(seed=42).select(range(100))\n", + " small_eval_dataset = tokenized_datasets[\"test\"].shuffle(seed=42).select(range(100))\n", + "\n", + " print(f\"len of train {small_train_dataset} and test {small_eval_dataset}\")\n", + "\n", + " ray_train_ds = ray.data.from_huggingface(small_train_dataset)\n", + " ray_evaluation_ds = ray.data.from_huggingface(small_eval_dataset)\n", + "\n", + " def compute_metrics(eval_pred):\n", + " metric = load_metric(\"accuracy\")\n", + " logits, labels = eval_pred\n", + " predictions = np.argmax(logits, axis=-1)\n", + " return metric.compute(predictions=predictions, references=labels)\n", + "\n", + " def trainer_init_per_worker(train_dataset, eval_dataset, **config):\n", + " model = AutoModelForSequenceClassification.from_pretrained(\"distilbert-base-uncased\", num_labels=2)\n", + "\n", + " training_args = TrainingArguments(\"/tmp/hf_imdb/test\", eval_steps=1, disable_tqdm=True, \n", + " num_train_epochs=1, skip_memory_metrics=True,\n", + " learning_rate=2e-5,\n", + " per_device_train_batch_size=16,\n", + " per_device_eval_batch_size=16, \n", + " weight_decay=0.01,)\n", + " return transformers.Trainer(\n", + " model=model,\n", + " args=training_args,\n", + " train_dataset=train_dataset,\n", + " eval_dataset=eval_dataset,\n", + " compute_metrics=compute_metrics\n", + " )\n", + "\n", + " scaling_config = ScalingConfig(num_workers=2, use_gpu=True) #num workers is the number of gpus\n", + "\n", + " # we are using the ray native HuggingFaceTrainer, but you can swap out to use non ray Huggingface Trainer. Both have the same method signature. \n", + " # the ray native HFTrainer has built in support for scaling to multiple GPUs\n", + " trainer = HuggingFaceTrainer(\n", + " trainer_init_per_worker=trainer_init_per_worker,\n", + " scaling_config=scaling_config,\n", + " datasets={\"train\": ray_train_ds, \"evaluation\": ray_evaluation_ds},\n", + " )\n", + " result = trainer.fit()" + ] + }, + { + "cell_type": "markdown", + "id": "d4d8fd65", + "metadata": {}, + "source": [ + "Once we want to test our code out, we can run the training function we defined above remotely on our Ray cluster:" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "5901d958", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Downloading builder script: 100%|██████████| 4.31k/4.31k [00:00<00:00, 4.22MB/s]\n", + "Downloading metadata: 100%|██████████| 2.17k/2.17k [00:00<00:00, 2.26MB/s]\n", + "Downloading readme: 100%|██████████| 7.59k/7.59k [00:00<00:00, 7.62MB/s]\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[2m\u001b[36m(train_fn pid=293)\u001b[0m Downloading and preparing dataset imdb/plain_text to /home/ray/.cache/huggingface/datasets/imdb/plain_text/1.0.0/d613c88cf8fa3bab83b4ded3713f1f74830d1100e171db75bbddb80b3345c9c0...\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Downloading data: 0%| | 0.00/84.1M [00:00 🚀 CodeFlare Cluster Details 🚀 \n", + " \n", + " ╭────────────────────────────────────────────────────────────────╮ \n", + " │ Name │ \n", + " │ gptfttest Active ✅ │ \n", + " │ │ \n", + " │ URI: ray://gptfttest-head-svc.default.svc:10001 │ \n", + " │ │ \n", + " │ Dashboard🔗 │ \n", + " │ │ \n", + " │ Cluster Resources │ \n", + " │ ╭─ Workers ──╮ ╭───────── Worker specs(each) ─────────╮ │ \n", + " │ │ Min Max │ │ Memory CPU GPU │ │ \n", + " │ │ │ │ │ │ \n", + " │ │ 2 2 │ │ 8~8 2 1 │ │ \n", + " │ │ │ │ │ │ \n", + " │ ╰────────────╯ ╰──────────────────────────────────────╯ │ \n", + " ╰────────────────────────────────────────────────────────────────╯ \n", + "\n" + ], + "text/plain": [ + "\u001b[3m \u001b[0m\u001b[1;3m 🚀 CodeFlare Cluster Details 🚀\u001b[0m\u001b[3m \u001b[0m\n", + "\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\n", + " ╭────────────────────────────────────────────────────────────────╮ \n", + " │ \u001b[1;37;42mName\u001b[0m │ \n", + " │ \u001b[1;4mgptfttest\u001b[0m Active ✅ │ \n", + " │ │ \n", + " │ \u001b[1mURI:\u001b[0m ray://gptfttest-head-svc.default.svc:10001 │ \n", + " │ │ \n", + " │ \u001b]8;id=618416;http://ray-dashboard-gptfttest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org\u001b\\\u001b[4;34mDashboard🔗\u001b[0m\u001b]8;;\u001b\\ │ \n", + " │ │ \n", + " │ \u001b[3m Cluster Resources \u001b[0m │ \n", + " │ ╭─ Workers ──╮ ╭───────── Worker specs(each) ─────────╮ │ \n", + " │ │ \u001b[1m \u001b[0m\u001b[1mMin\u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mMax\u001b[0m\u001b[1m \u001b[0m │ │ \u001b[1m \u001b[0m\u001b[1mMemory \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mCPU \u001b[0m\u001b[1m \u001b[0m\u001b[1m \u001b[0m\u001b[1mGPU \u001b[0m\u001b[1m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m2 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m2 \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m8~8 \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m2 \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m1 \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[36m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m\u001b[35m \u001b[0m │ │ \n", + " │ ╰────────────╯ ╰──────────────────────────────────────╯ │ \n", + " ╰────────────────────────────────────────────────────────────────╯ \n" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "text/plain": [ + "RayCluster(name='gptfttest', status=, min_workers=2, max_workers=2, worker_mem_min=8, worker_mem_max=8, worker_cpu=2, worker_gpu=1, namespace='default', dashboard='http://ray-dashboard-gptfttest-default.apps.meyceoz-032023.psap.aws.rhperfscale.org')" + ] + }, + "execution_count": 662, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "cluster.details()" + ] + }, + { + "cell_type": "code", + "execution_count": 663, + "id": "9ed5bd75-4230-4c7c-a9e2-0f247890e62a", + "metadata": {}, + "outputs": [], + "source": [ + "from codeflare_sdk.job.jobs import DDPJobDefinition" + ] + }, + { + "cell_type": "code", + "execution_count": 664, + "id": "611d203a-35aa-4357-a748-1d01b022fcdb", + "metadata": {}, + "outputs": [], + "source": [ + "arg_list = [\n", + " \"--model_name_or_path\", \"gpt2\",\n", + " \"--dataset_name\", \"wikitext\",\n", + " \"--dataset_config_name\", \"wikitext-2-raw-v1\",\n", + " \"--per_device_train_batch_size\", \"2\",\n", + " \"--per_device_eval_batch_size\", \"2\",\n", + " \"--do_train\",\n", + " \"--do_eval\",\n", + " \"--output_dir\", \"/tmp/test-clm\",\n", + " \"--overwrite_output_dir\"\n", + "]" + ] + }, + { + "cell_type": "code", + "execution_count": 665, + "id": "8ac7c34f-e227-44c2-a4b1-a57c853ac3a7", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "The Ray scheduler does not support port mapping.\n" + ] + } + ], + "source": [ + "jobdef = DDPJobDefinition(\n", + " name=\"gpttest\",\n", + " script=\"gpt_og.py\",\n", + " script_args=arg_list,\n", + " scheduler_args={\"requirements\": \"requirements_gpt.txt\"}\n", + ")\n", + "job = jobdef.submit(cluster)" + ] + }, + { + "cell_type": "code", + "execution_count": 675, + "id": "1680d287-de46-45f8-b95a-02ba3c83912c", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "AppStatus:\n", + " msg: !!python/object/apply:ray.dashboard.modules.job.common.JobStatus\n", + " - SUCCEEDED\n", + " num_restarts: -1\n", + " roles:\n", + " - replicas:\n", + " - hostname: \n", + " id: 0\n", + " role: ray\n", + " state: !!python/object/apply:torchx.specs.api.AppState\n", + " - 4\n", + " structured_error_msg: \n", + " role: ray\n", + " state: SUCCEEDED (4)\n", + " structured_error_msg: \n", + " ui_url: null" + ] + }, + "execution_count": 675, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "job.status()" + ] + }, + { + "cell_type": "code", + "execution_count": 674, + "id": "d25d6198-9941-47e8-857f-9811830cc854", + "metadata": {}, + "outputs": [ + { + "data": { + "text/plain": [ + "'[RayActor(name=\\'gpt_og\\', command=[\\'bash\\', \\'-c\\', \"torchrun --rdzv_backend static --rdzv_endpoint $TORCHX_RANK0_HOST:49782 --rdzv_id \\'gpttest-czbl5zhq5hnl7c\\' --nnodes 2 --nproc_per_node 1 --node_rank \\'0\\' --tee 3 --role \\'\\' gpt_og.py --model_name_or_path gpt2 --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --per_device_train_batch_size 2 --per_device_eval_batch_size 2 --do_train --do_eval --output_dir /tmp/test-clm --overwrite_output_dir\"], env={\\'TORCHX_TRACKING_EXPERIMENT_NAME\\': \\'default-experiment\\', \\'LOGLEVEL\\': \\'WARNING\\', \\'TORCHX_JOB_ID\\': \\'ray://torchx/gpttest-czbl5zhq5hnl7c\\'}, num_cpus=2, num_gpus=1, min_replicas=2), RayActor(name=\\'gpt_og\\', command=[\\'bash\\', \\'-c\\', \"torchrun --rdzv_backend static --rdzv_endpoint $TORCHX_RANK0_HOST:49782 --rdzv_id \\'gpttest-czbl5zhq5hnl7c\\' --nnodes 2 --nproc_per_node 1 --node_rank \\'1\\' --tee 3 --role \\'\\' gpt_og.py --model_name_or_path gpt2 --dataset_name wikitext --dataset_config_name wikitext-2-raw-v1 --per_device_train_batch_size 2 --per_device_eval_batch_size 2 --do_train --do_eval --output_dir /tmp/test-clm --overwrite_output_dir\"], env={\\'TORCHX_TRACKING_EXPERIMENT_NAME\\': \\'default-experiment\\', \\'LOGLEVEL\\': \\'WARNING\\', \\'TORCHX_JOB_ID\\': \\'ray://torchx/gpttest-czbl5zhq5hnl7c\\'}, num_cpus=2, num_gpus=1, min_replicas=2)]\\n2023-04-26 07:31:11,126\\tINFO worker.py:1230 -- Using address 10.131.20.8:6379 set in the environment variable RAY_ADDRESS\\n2023-04-26 07:31:11,126\\tINFO worker.py:1342 -- Connecting to existing Ray cluster at address: 10.131.20.8:6379...\\n2023-04-26 07:31:11,133\\tINFO worker.py:1519 -- Connected to Ray cluster. View the dashboard at \\x1b[1m\\x1b[32mhttp://10.131.20.8:8265 \\x1b[39m\\x1b[22m\\nWaiting for minimum placement group to start.\\nSuccessfully created placement groups\\nrdzv_endpoint set to 10.129.22.18 for actor 937fc115f41dd3f37cce443a02000000\\nrdzv_endpoint set to 10.129.22.18 for actor 9c109cdc7c9e0dbb9d916c6702000000\\nSuccessfully placed command actors\\nEntering main loop, start executing the script on worker nodes\\nrunning ray.wait on [ObjectRef(e082c90ab8422b00937fc115f41dd3f37cce443a0200000001000000), ObjectRef(ce868e48e2fa9a949c109cdc7c9e0dbb9d916c670200000001000000)]\\nrunning ray.wait on [ObjectRef(ce868e48e2fa9a949c109cdc7c9e0dbb9d916c670200000001000000), ObjectRef(f81ec6ff838b16db937fc115f41dd3f37cce443a0200000001000000)]\\nrunning ray.wait on [ObjectRef(f81ec6ff838b16db937fc115f41dd3f37cce443a0200000001000000), ObjectRef(32b0eec39cfa87ac9c109cdc7c9e0dbb9d916c670200000001000000)]\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Found existing installation: protobuf 3.19.6\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Not uninstalling protobuf at /home/ray/anaconda3/lib/python3.8/site-packages, outside environment /tmp/ray/session_2023-04-26_07-13-38_751985_7/runtime_resources/pip/98fab9b6ed3822d73f034e53208a68b433de183d/virtualenv\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Can\\'t uninstall \\'protobuf\\'. No files were found to uninstall.\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting accelerate>=0.12.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading accelerate-0.18.0-py3-none-any.whl (215 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 215.3/215.3 kB 39.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting torch>=1.3\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading torch-2.0.0-cp38-cp38-manylinux1_x86_64.whl (619.9 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 619.9/619.9 MB 7.5 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting datasets>=1.8.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading datasets-2.11.0-py3-none-any.whl (468 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 468.7/468.7 kB 125.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting sentencepiece!=0.1.92\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading sentencepiece-0.1.98-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.3/1.3 MB 144.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting evaluate\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading evaluate-0.4.0-py3-none-any.whl (81 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 81.4/81.4 kB 54.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting scikit-learn\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading scikit_learn-1.2.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (9.8 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 9.8/9.8 MB 121.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting transformers==4.28.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading transformers-4.28.1-py3-none-any.whl (7.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7.0/7.0 MB 99.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting protobuf<=3.20.1,>=3.8.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading protobuf-3.20.1-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.whl (1.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.0/1.0 MB 131.4 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting regex!=2019.12.17\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading regex-2023.3.23-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (771 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 771.9/771.9 kB 159.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting packaging>=20.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading packaging-23.1-py3-none-any.whl (48 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 48.9/48.9 kB 35.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting tqdm>=4.27\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading tqdm-4.65.0-py3-none-any.whl (77 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 77.1/77.1 kB 48.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting numpy>=1.17\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading numpy-1.24.3-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (17.3 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 17.3/17.3 MB 141.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting filelock\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading filelock-3.12.0-py3-none-any.whl (10 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting huggingface-hub<1.0,>=0.11.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading huggingface_hub-0.14.1-py3-none-any.whl (224 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 224.5/224.5 kB 103.1 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting pyyaml>=5.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading PyYAML-6.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (701 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 701.2/701.2 kB 39.6 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting requests\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading requests-2.28.2-py3-none-any.whl (62 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 62.8/62.8 kB 48.5 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting tokenizers!=0.11.3,<0.14,>=0.11.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading tokenizers-0.13.3-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.8 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7.8/7.8 MB 160.4 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting psutil\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading psutil-5.9.5-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (282 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 282.1/282.1 kB 121.5 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-cudnn-cu11==8.5.0.96\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_cudnn_cu11-8.5.0.96-2-py3-none-manylinux1_x86_64.whl (557.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 557.1/557.1 MB 6.5 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting sympy\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading sympy-1.11.1-py3-none-any.whl (6.5 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.5/6.5 MB 149.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-cuda-cupti-cu11==11.7.101\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_cuda_cupti_cu11-11.7.101-py3-none-manylinux1_x86_64.whl (11.8 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 11.8/11.8 MB 136.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting jinja2\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading Jinja2-3.1.2-py3-none-any.whl (133 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 133.1/133.1 kB 73.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting triton==2.0.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading triton-2.0.0-1-cp38-cp38-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (63.2 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 63.2/63.2 MB 57.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-cuda-nvrtc-cu11==11.7.99\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_cuda_nvrtc_cu11-11.7.99-2-py3-none-manylinux1_x86_64.whl (21.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 21.0/21.0 MB 132.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-nvtx-cu11==11.7.91\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_nvtx_cu11-11.7.91-py3-none-manylinux1_x86_64.whl (98 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 98.6/98.6 kB 59.4 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-nccl-cu11==2.14.3\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_nccl_cu11-2.14.3-py3-none-manylinux1_x86_64.whl (177.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 177.1/177.1 MB 24.7 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-cusolver-cu11==11.4.0.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_cusolver_cu11-11.4.0.1-2-py3-none-manylinux1_x86_64.whl (102.6 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 102.6/102.6 MB 39.7 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-curand-cu11==10.2.10.91\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_curand_cu11-10.2.10.91-py3-none-manylinux1_x86_64.whl (54.6 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 54.6/54.6 MB 61.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting networkx\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading networkx-3.1-py3-none-any.whl (2.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Found existing installation: protobuf 3.19.6\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Not uninstalling protobuf at /home/ray/anaconda3/lib/python3.8/site-packages, outside environment /tmp/ray/session_2023-04-26_07-13-38_751985_7/runtime_resources/pip/98fab9b6ed3822d73f034e53208a68b433de183d/virtualenv\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Can\\'t uninstall \\'protobuf\\'. No files were found to uninstall.\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting accelerate>=0.12.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading accelerate-0.18.0-py3-none-any.whl (215 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 215.3/215.3 kB 36.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting torch>=1.3\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading torch-2.0.0-cp38-cp38-manylinux1_x86_64.whl (619.9 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 619.9/619.9 MB 7.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting datasets>=1.8.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading datasets-2.11.0-py3-none-any.whl (468 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 468.7/468.7 kB 123.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting sentencepiece!=0.1.92\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading sentencepiece-0.1.98-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.3/1.3 MB 143.5 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting evaluate\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading evaluate-0.4.0-py3-none-any.whl (81 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 81.4/81.4 kB 53.5 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting scikit-learn\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading scikit_learn-1.2.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (9.8 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 9.8/9.8 MB 150.7 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting transformers==4.28.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading transformers-4.28.1-py3-none-any.whl (7.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7.0/7.0 MB 152.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting protobuf<=3.20.1,>=3.8.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading protobuf-3.20.1-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.whl (1.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.0/1.0 MB 166.1 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting numpy>=1.17\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading numpy-1.24.3-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (17.3 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 17.3/17.3 MB 117.5 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting requests\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading requests-2.28.2-py3-none-any.whl (62 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 62.8/62.8 kB 42.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting filelock\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading filelock-3.12.0-py3-none-any.whl (10 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting packaging>=20.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading packaging-23.1-py3-none-any.whl (48 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 48.9/48.9 kB 33.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting pyyaml>=5.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading PyYAML-6.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_12_x86_64.manylinux2010_x86_64.whl (701 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 701.2/701.2 kB 149.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting tokenizers!=0.11.3,<0.14,>=0.11.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading tokenizers-0.13.3-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (7.8 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7.8/7.8 MB 162.4 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting tqdm>=4.27\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading tqdm-4.65.0-py3-none-any.whl (77 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 77.1/77.1 kB 48.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting huggingface-hub<1.0,>=0.11.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading huggingface_hub-0.14.1-py3-none-any.whl (224 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 224.5/224.5 kB 104.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting regex!=2019.12.17\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading regex-2023.3.23-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (771 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 771.9/771.9 kB 157.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting psutil\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading psutil-5.9.5-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (282 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 282.1/282.1 kB 115.6 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-cusparse-cu11==11.7.4.91\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_cusparse_cu11-11.7.4.91-py3-none-manylinux1_x86_64.whl (173.2 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 173.2/173.2 MB 23.6 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting triton==2.0.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading triton-2.0.0-1-cp38-cp38-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (63.2 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 63.2/63.2 MB 54.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-cublas-cu11==11.10.3.66\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_cublas_cu11-11.10.3.66-py3-none-manylinux1_x86_64.whl (317.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 317.1/317.1 MB 14.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-cufft-cu11==10.9.0.58\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_cufft_cu11-10.9.0.58-py3-none-manylinux1_x86_64.whl (168.4 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 168.4/168.4 MB 24.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-nccl-cu11==2.14.3\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_nccl_cu11-2.14.3-py3-none-manylinux1_x86_64.whl (177.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 177.1/177.1 MB 24.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting sympy\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading sympy-1.11.1-py3-none-any.whl (6.5 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.5/6.5 MB 168.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-cudnn-cu11==8.5.0.96\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_cudnn_cu11-8.5.0.96-2-py3-none-manylinux1_x86_64.whl (557.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 557.1/557.1 MB 3.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-curand-cu11==10.2.10.91\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_curand_cu11-10.2.10.91-py3-none-manylinux1_x86_64.whl (54.6 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 54.6/54.6 MB 66.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting jinja2\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading Jinja2-3.1.2-py3-none-any.whl (133 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 133.1/133.1 kB 73.7 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting networkx\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading networkx-3.1-py3-none-any.whl (2.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.1/2.1 MB 162.6 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-nvtx-cu11==11.7.91\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_nvtx_cu11-11.7.91-py3-none-manylinux1_x86_64.whl (98 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.1/2.1 MB 177.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-cublas-cu11==11.10.3.66\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_cublas_cu11-11.10.3.66-py3-none-manylinux1_x86_64.whl (317.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 317.1/317.1 MB 13.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-cufft-cu11==10.9.0.58\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_cufft_cu11-10.9.0.58-py3-none-manylinux1_x86_64.whl (168.4 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 168.4/168.4 MB 25.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting typing-extensions\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading typing_extensions-4.5.0-py3-none-any.whl (27 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-cusparse-cu11==11.7.4.91\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_cusparse_cu11-11.7.4.91-py3-none-manylinux1_x86_64.whl (173.2 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 173.2/173.2 MB 23.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting nvidia-cuda-runtime-cu11==11.7.99\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading nvidia_cuda_runtime_cu11-11.7.99-py3-none-manylinux1_x86_64.whl (849 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 849.3/849.3 kB 162.7 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting wheel\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading wheel-0.40.0-py3-none-any.whl (64 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 64.5/64.5 kB 47.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting setuptools\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading setuptools-67.7.2-py3-none-any.whl (1.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.1/1.1 MB 164.4 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting lit\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading lit-16.0.2.tar.gz (137 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 137.9/137.9 kB 83.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Preparing metadata (setup.py): started\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Preparing metadata (setup.py): finished with status \\'done\\'\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting cmake\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading cmake-3.26.3-py2.py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (24.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 24.0/24.0 MB 109.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting fsspec[http]>=2021.11.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading fsspec-2023.4.0-py3-none-any.whl (153 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 154.0/154.0 kB 90.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting pyarrow>=8.0.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading pyarrow-11.0.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (35.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 35.0/35.0 MB 86.1 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting pandas\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading pandas-2.0.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (12.3 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 12.3/12.3 MB 150.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting multiprocess\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading multiprocess-0.70.14-py38-none-any.whl (132 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 132.0/132.0 kB 80.7 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting responses<0.19\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading responses-0.18.0-py3-none-any.whl (38 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting aiohttp\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading aiohttp-3.8.4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.0/1.0 MB 172.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting xxhash\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading xxhash-3.2.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (213 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 213.0/213.0 kB 107.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting dill<0.3.7,>=0.3.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading dill-0.3.6-py3-none-any.whl (110 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 110.5/110.5 kB 45.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting threadpoolctl>=2.0.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading threadpoolctl-3.1.0-py3-none-any.whl (14 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting joblib>=1.1.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading joblib-1.2.0-py3-none-any.whl (297 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 298.0/298.0 kB 118.1 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting scipy>=1.3.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading scipy-1.10.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (34.5 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 34.5/34.5 MB 81.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting attrs>=17.3.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading attrs-23.1.0-py3-none-any.whl (61 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.2/61.2 kB 45.5 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting charset-normalizer<4.0,>=2.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading charset_normalizer-3.1.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (195 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 195.9/195.9 kB 99.1 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting multidict<7.0,>=4.5\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading multidict-6.0.4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (121 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 121.3/121.3 kB 73.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting frozenlist>=1.1.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading frozenlist-1.3.3-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (161 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 161.3/161.3 kB 90.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting async-timeout<5.0,>=4.0.0a3\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading async_timeout-4.0.2-py3-none-any.whl (5.8 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting yarl<2.0,>=1.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading yarl-1.9.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (266 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 266.9/266.9 kB 113.6 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting aiosignal>=1.1.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading aiosignal-1.3.1-py3-none-any.whl (7.6 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting certifi>=2017.4.17\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading certifi-2022.12.7-py3-none-any.whl (155 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 155.3/155.3 kB 87.4 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting urllib3<1.27,>=1.21.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading urllib3-1.26.15-py2.py3-none-any.whl (140 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 140.9/140.9 kB 81.1 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting idna<4,>=2.5\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading idna-3.4-py3-none-any.whl (61 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.5/61.5 kB 43.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting MarkupSafe>=2.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading MarkupSafe-2.1.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (25 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting python-dateutil>=2.8.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading python_dateutil-2.8.2-py2.py3-none-any.whl (247 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 247.7/247.7 kB 112.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting tzdata>=2022.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading tzdata-2023.3-py2.py3-none-any.whl (341 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 98.6/98.6 kB 62.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-cuda-nvrtc-cu11==11.7.99\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_cuda_nvrtc_cu11-11.7.99-2-py3-none-manylinux1_x86_64.whl (21.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 21.0/21.0 MB 114.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-cusolver-cu11==11.4.0.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_cusolver_cu11-11.4.0.1-2-py3-none-manylinux1_x86_64.whl (102.6 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 102.6/102.6 MB 39.6 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-cuda-runtime-cu11==11.7.99\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_cuda_runtime_cu11-11.7.99-py3-none-manylinux1_x86_64.whl (849 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 849.3/849.3 kB 146.6 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting typing-extensions\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading typing_extensions-4.5.0-py3-none-any.whl (27 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting nvidia-cuda-cupti-cu11==11.7.101\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading nvidia_cuda_cupti_cu11-11.7.101-py3-none-manylinux1_x86_64.whl (11.8 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 11.8/11.8 MB 167.6 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting wheel\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading wheel-0.40.0-py3-none-any.whl (64 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 64.5/64.5 kB 43.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting setuptools\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading setuptools-67.7.2-py3-none-any.whl (1.1 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.1/1.1 MB 160.1 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting cmake\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading cmake-3.26.3-py2.py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (24.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 24.0/24.0 MB 119.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting lit\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading lit-16.0.2.tar.gz (137 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 137.9/137.9 kB 80.8 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Preparing metadata (setup.py): started\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Preparing metadata (setup.py): finished with status \\'done\\'\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting aiohttp\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading aiohttp-3.8.4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.0/1.0 MB 163.6 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting pyarrow>=8.0.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading pyarrow-11.0.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (35.0 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 35.0/35.0 MB 93.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting dill<0.3.7,>=0.3.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading dill-0.3.6-py3-none-any.whl (110 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 110.5/110.5 kB 71.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting multiprocess\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading multiprocess-0.70.14-py38-none-any.whl (132 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 132.0/132.0 kB 73.9 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting xxhash\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading xxhash-3.2.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (213 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 213.0/213.0 kB 102.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting fsspec[http]>=2021.11.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading fsspec-2023.4.0-py3-none-any.whl (153 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 154.0/154.0 kB 89.7 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting responses<0.19\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading responses-0.18.0-py3-none-any.whl (38 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting pandas\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading pandas-2.0.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (12.3 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 12.3/12.3 MB 139.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting joblib>=1.1.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading joblib-1.2.0-py3-none-any.whl (297 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 298.0/298.0 kB 119.1 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting threadpoolctl>=2.0.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading threadpoolctl-3.1.0-py3-none-any.whl (14 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting scipy>=1.3.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading scipy-1.10.1-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (34.5 MB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 34.5/34.5 MB 86.4 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting charset-normalizer<4.0,>=2.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading charset_normalizer-3.1.0-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (195 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 195.9/195.9 kB 100.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting frozenlist>=1.1.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading frozenlist-1.3.3-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (161 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 161.3/161.3 kB 88.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting aiosignal>=1.1.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading aiosignal-1.3.1-py3-none-any.whl (7.6 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting attrs>=17.3.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading attrs-23.1.0-py3-none-any.whl (61 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.2/61.2 kB 39.5 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting yarl<2.0,>=1.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading yarl-1.9.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (266 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 266.9/266.9 kB 113.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting multidict<7.0,>=4.5\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading multidict-6.0.4-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (121 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 121.3/121.3 kB 69.1 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting async-timeout<5.0,>=4.0.0a3\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading async_timeout-4.0.2-py3-none-any.whl (5.8 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting certifi>=2017.4.17\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading certifi-2022.12.7-py3-none-any.whl (155 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 155.3/155.3 kB 86.7 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting idna<4,>=2.5\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading idna-3.4-py3-none-any.whl (61 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.5/61.5 kB 41.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting urllib3<1.27,>=1.21.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading urllib3-1.26.15-py2.py3-none-any.whl (140 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 140.9/140.9 kB 79.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting MarkupSafe>=2.0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading MarkupSafe-2.1.2-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (25 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting pytz>=2020.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading pytz-2023.3-py2.py3-none-any.whl (502 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 502.3/502.3 kB 148.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting python-dateutil>=2.8.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading python_dateutil-2.8.2-py2.py3-none-any.whl (247 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 341.8/341.8 kB 126.0 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting pytz>=2020.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading pytz-2023.3-py2.py3-none-any.whl (502 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 502.3/502.3 kB 135.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting mpmath>=0.19\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading mpmath-1.3.0-py3-none-any.whl (536 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 536.2/536.2 kB 144.7 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Collecting six>=1.5\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Downloading six-1.16.0-py2.py3-none-any.whl (11 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Building wheels for collected packages: lit\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Building wheel for lit (setup.py): started\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Building wheel for lit (setup.py): finished with status \\'done\\'\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Created wheel for lit: filename=lit-16.0.2-py3-none-any.whl size=88174 sha256=b5e3a34a26c7dff85807f4525c1e1713d7ef47e58d8312d3a82495d1d2ffbc61\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: Stored in directory: /home/ray/.cache/pip/wheels/db/ec/82/32d94d2fa45c0af452689a6c91046392b29d856fe2eebc8276\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Successfully built lit\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Installing collected packages: tokenizers, sentencepiece, pytz, mpmath, lit, cmake, xxhash, wheel, urllib3, tzdata, typing-extensions, tqdm, threadpoolctl, sympy, six, setuptools, regex, pyyaml, psutil, protobuf, packaging, nvidia-nccl-cu11, nvidia-cufft-cu11, nvidia-cuda-nvrtc-cu11, numpy, networkx, multidict, MarkupSafe, joblib, idna, fsspec, frozenlist, filelock, dill, charset-normalizer, certifi, attrs, async-timeout, yarl, scipy, requests, python-dateutil, pyarrow, nvidia-nvtx-cu11, nvidia-cusparse-cu11, nvidia-curand-cu11, nvidia-cuda-runtime-cu11, nvidia-cuda-cupti-cu11, nvidia-cublas-cu11, multiprocess, jinja2, aiosignal, scikit-learn, responses, pandas, nvidia-cusolver-cu11, nvidia-cudnn-cu11, huggingface-hub, aiohttp, transformers, datasets, evaluate, triton, torch, accelerate\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Successfully installed MarkupSafe-2.1.2 accelerate-0.18.0 aiohttp-3.8.4 aiosignal-1.3.1 async-timeout-4.0.2 attrs-23.1.0 certifi-2022.12.7 charset-normalizer-3.1.0 cmake-3.26.3 datasets-2.11.0 dill-0.3.6 evaluate-0.4.0 filelock-3.12.0 frozenlist-1.3.3 fsspec-2023.4.0 huggingface-hub-0.14.1 idna-3.4 jinja2-3.1.2 joblib-1.2.0 lit-16.0.2 mpmath-1.3.0 multidict-6.0.4 multiprocess-0.70.14 networkx-3.1 numpy-1.24.3 nvidia-cublas-cu11-11.10.3.66 nvidia-cuda-cupti-cu11-11.7.101 nvidia-cuda-nvrtc-cu11-11.7.99 nvidia-cuda-runtime-cu11-11.7.99 nvidia-cudnn-cu11-8.5.0.96 nvidia-cufft-cu11-10.9.0.58 nvidia-curand-cu11-10.2.10.91 nvidia-cusolver-cu11-11.4.0.1 nvidia-cusparse-cu11-11.7.4.91 nvidia-nccl-cu11-2.14.3 nvidia-nvtx-cu11-11.7.91 packaging-23.1 pandas-2.0.1 protobuf-3.20.1 psutil-5.9.5 pyarrow-11.0.0 python-dateutil-2.8.2 pytz-2023.3 pyyaml-6.0 regex-2023.3.23 requests-2.28.2 responses-0.18.0 scikit-learn-1.2.2 scipy-1.10.1 sentencepiece-0.1.98 setuptools-67.7.2 six-1.16.0 sympy-1.11.1 threadpoolctl-3.1.0 tokenizers-0.13.3 torch-2.0.0 tqdm-4.65.0 transformers-4.28.1 triton-2.0.0 typing-extensions-4.5.0 tzdata-2023.3 urllib3-1.26.15 wheel-0.40.0 xxhash-3.2.0 yarl-1.9.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:04/26/2023 07:33:24 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1distributed training: True, 16-bits training: False\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:04/26/2023 07:33:24 - INFO - __main__ - Training/evaluation parameters TrainingArguments(\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:_n_gpu=1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:adafactor=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:adam_beta1=0.9,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:adam_beta2=0.999,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:adam_epsilon=1e-08,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:auto_find_batch_size=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:bf16=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:bf16_full_eval=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:data_seed=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:dataloader_drop_last=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:dataloader_num_workers=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:dataloader_pin_memory=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:ddp_bucket_cap_mb=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:ddp_find_unused_parameters=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:ddp_timeout=1800,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:debug=[],\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:deepspeed=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:disable_tqdm=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:do_eval=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:do_predict=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:do_train=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:eval_accumulation_steps=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:eval_delay=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:eval_steps=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:evaluation_strategy=no,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:fp16=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:fp16_backend=auto,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:fp16_full_eval=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:fp16_opt_level=O1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:fsdp=[],\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:fsdp_config={\\'fsdp_min_num_params\\': 0, \\'xla\\': False, \\'xla_fsdp_grad_ckpt\\': False},\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:fsdp_min_num_params=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:fsdp_transformer_layer_cls_to_wrap=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:full_determinism=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:gradient_accumulation_steps=1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:gradient_checkpointing=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:greater_is_better=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:group_by_length=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:half_precision_backend=auto,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:hub_model_id=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:hub_private_repo=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:hub_strategy=every_save,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:hub_token=,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:ignore_data_skip=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:include_inputs_for_metrics=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:jit_mode_eval=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:label_names=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:label_smoothing_factor=0.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:learning_rate=5e-05,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:length_column_name=length,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:load_best_model_at_end=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:local_rank=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:log_level=passive,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:log_level_replica=warning,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:log_on_each_node=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:logging_dir=/tmp/test-clm/runs/Apr26_07-33-20_gptfttest-worker-small-group-gptfttest-7fdjz,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:logging_first_step=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:logging_nan_inf_filter=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:logging_steps=500,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:logging_strategy=steps,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:lr_scheduler_type=linear,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:max_grad_norm=1.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:max_steps=-1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:metric_for_best_model=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:mp_parameters=,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:no_cuda=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:num_train_epochs=3.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:optim=adamw_hf,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:optim_args=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:output_dir=/tmp/test-clm,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:overwrite_output_dir=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:past_index=-1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:per_device_eval_batch_size=2,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:per_device_train_batch_size=2,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:prediction_loss_only=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:push_to_hub=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:push_to_hub_model_id=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:push_to_hub_organization=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:push_to_hub_token=,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:ray_scope=last,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:remove_unused_columns=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:report_to=[\\'tensorboard\\'],\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:resume_from_checkpoint=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:run_name=/tmp/test-clm,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:save_on_each_node=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:save_safetensors=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:save_steps=500,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:save_strategy=steps,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:save_total_limit=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:seed=42,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:sharded_ddp=[],\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:skip_memory_metrics=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:tf32=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:torch_compile=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:torch_compile_backend=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:torch_compile_mode=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:torchdynamo=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:tpu_metrics_debug=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:tpu_num_cores=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:use_ipex=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:use_legacy_prediction_loop=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:use_mps_device=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:warmup_ratio=0.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:warmup_steps=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:weight_decay=0.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:xpu_backend=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:)\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - https://huggingface.co/datasets/wikitext/resolve/main/wikitext.py not found in cache or force_download set to True, downloading to /home/ray/.cache/huggingface/datasets/downloads/tmp9__iv6ze\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - storing https://huggingface.co/datasets/wikitext/resolve/main/wikitext.py in cache at /home/ray/.cache/huggingface/datasets/downloads/30cb21e192e211952c02572882251280460fb5247fe18b6c0fb69224e769f1e1.6a998136b3179c543fac19963253d25970e7fe6d053f2818edc7075627f64bad.py\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - creating metadata file for /home/ray/.cache/huggingface/datasets/downloads/30cb21e192e211952c02572882251280460fb5247fe18b6c0fb69224e769f1e1.6a998136b3179c543fac19963253d25970e7fe6d053f2818edc7075627f64bad.py\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - https://huggingface.co/datasets/wikitext/resolve/main/dataset_infos.json not found in cache or force_download set to True, downloading to /home/ray/.cache/huggingface/datasets/downloads/tmp0_knks6x\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - storing https://huggingface.co/datasets/wikitext/resolve/main/dataset_infos.json in cache at /home/ray/.cache/huggingface/datasets/downloads/87ea4775c52b60feb08a5087c68f4453d4533a02491172390b4d6a3f97ae44d1.d3aa47a864d0b5cf3b7ebcf51e45c9d8f96356ff8527fff02d3a4cae4c9f5b1e\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - creating metadata file for /home/ray/.cache/huggingface/datasets/downloads/87ea4775c52b60feb08a5087c68f4453d4533a02491172390b4d6a3f97ae44d1.d3aa47a864d0b5cf3b7ebcf51e45c9d8f96356ff8527fff02d3a4cae4c9f5b1e\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 247.7/247.7 kB 116.2 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting tzdata>=2022.1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading tzdata-2023.3-py2.py3-none-any.whl (341 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 341.8/341.8 kB 133.3 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting mpmath>=0.19\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading mpmath-1.3.0-py3-none-any.whl (536 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 536.2/536.2 kB 150.4 MB/s eta 0:00:00\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Collecting six>=1.5\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Downloading six-1.16.0-py2.py3-none-any.whl (11 kB)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Building wheels for collected packages: lit\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Building wheel for lit (setup.py): started\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Building wheel for lit (setup.py): finished with status \\'done\\'\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Created wheel for lit: filename=lit-16.0.2-py3-none-any.whl size=88174 sha256=b5e3a34a26c7dff85807f4525c1e1713d7ef47e58d8312d3a82495d1d2ffbc61\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: Stored in directory: /home/ray/.cache/pip/wheels/db/ec/82/32d94d2fa45c0af452689a6c91046392b29d856fe2eebc8276\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Successfully built lit\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Installing collected packages: tokenizers, sentencepiece, pytz, mpmath, lit, cmake, xxhash, wheel, urllib3, tzdata, typing-extensions, tqdm, threadpoolctl, sympy, six, setuptools, regex, pyyaml, psutil, protobuf, packaging, nvidia-nccl-cu11, nvidia-cufft-cu11, nvidia-cuda-nvrtc-cu11, numpy, networkx, multidict, MarkupSafe, joblib, idna, fsspec, frozenlist, filelock, dill, charset-normalizer, certifi, attrs, async-timeout, yarl, scipy, requests, python-dateutil, pyarrow, nvidia-nvtx-cu11, nvidia-cusparse-cu11, nvidia-curand-cu11, nvidia-cuda-runtime-cu11, nvidia-cuda-cupti-cu11, nvidia-cublas-cu11, multiprocess, jinja2, aiosignal, scikit-learn, responses, pandas, nvidia-cusolver-cu11, nvidia-cudnn-cu11, huggingface-hub, aiohttp, transformers, datasets, evaluate, triton, torch, accelerate\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Successfully installed MarkupSafe-2.1.2 accelerate-0.18.0 aiohttp-3.8.4 aiosignal-1.3.1 async-timeout-4.0.2 attrs-23.1.0 certifi-2022.12.7 charset-normalizer-3.1.0 cmake-3.26.3 datasets-2.11.0 dill-0.3.6 evaluate-0.4.0 filelock-3.12.0 frozenlist-1.3.3 fsspec-2023.4.0 huggingface-hub-0.14.1 idna-3.4 jinja2-3.1.2 joblib-1.2.0 lit-16.0.2 mpmath-1.3.0 multidict-6.0.4 multiprocess-0.70.14 networkx-3.1 numpy-1.24.3 nvidia-cublas-cu11-11.10.3.66 nvidia-cuda-cupti-cu11-11.7.101 nvidia-cuda-nvrtc-cu11-11.7.99 nvidia-cuda-runtime-cu11-11.7.99 nvidia-cudnn-cu11-8.5.0.96 nvidia-cufft-cu11-10.9.0.58 nvidia-curand-cu11-10.2.10.91 nvidia-cusolver-cu11-11.4.0.1 nvidia-cusparse-cu11-11.7.4.91 nvidia-nccl-cu11-2.14.3 nvidia-nvtx-cu11-11.7.91 packaging-23.1 pandas-2.0.1 protobuf-3.20.1 psutil-5.9.5 pyarrow-11.0.0 python-dateutil-2.8.2 pytz-2023.3 pyyaml-6.0 regex-2023.3.23 requests-2.28.2 responses-0.18.0 scikit-learn-1.2.2 scipy-1.10.1 sentencepiece-0.1.98 setuptools-67.7.2 six-1.16.0 sympy-1.11.1 threadpoolctl-3.1.0 tokenizers-0.13.3 torch-2.0.0 tqdm-4.65.0 transformers-4.28.1 triton-2.0.0 typing-extensions-4.5.0 tzdata-2023.3 urllib3-1.26.15 wheel-0.40.0 xxhash-3.2.0 yarl-1.9.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:04/26/2023 07:33:24 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1distributed training: True, 16-bits training: False\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:04/26/2023 07:33:24 - INFO - __main__ - Training/evaluation parameters TrainingArguments(\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:_n_gpu=1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:adafactor=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:adam_beta1=0.9,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:adam_beta2=0.999,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:adam_epsilon=1e-08,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:auto_find_batch_size=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:bf16=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:bf16_full_eval=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:data_seed=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:dataloader_drop_last=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:dataloader_num_workers=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:dataloader_pin_memory=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:ddp_bucket_cap_mb=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:ddp_find_unused_parameters=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:ddp_timeout=1800,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:debug=[],\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:deepspeed=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:disable_tqdm=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:do_eval=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:do_predict=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:do_train=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:eval_accumulation_steps=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:eval_delay=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:eval_steps=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:evaluation_strategy=no,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:fp16=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:fp16_backend=auto,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:fp16_full_eval=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:fp16_opt_level=O1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:fsdp=[],\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:fsdp_config={\\'fsdp_min_num_params\\': 0, \\'xla\\': False, \\'xla_fsdp_grad_ckpt\\': False},\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:fsdp_min_num_params=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:fsdp_transformer_layer_cls_to_wrap=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:full_determinism=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:gradient_accumulation_steps=1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:gradient_checkpointing=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:greater_is_better=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:group_by_length=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:half_precision_backend=auto,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:hub_model_id=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:hub_private_repo=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:hub_strategy=every_save,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:hub_token=,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:ignore_data_skip=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:include_inputs_for_metrics=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:jit_mode_eval=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:label_names=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:label_smoothing_factor=0.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:learning_rate=5e-05,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:length_column_name=length,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:load_best_model_at_end=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:local_rank=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:log_level=passive,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:log_level_replica=warning,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:log_on_each_node=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:logging_dir=/tmp/test-clm/runs/Apr26_07-33-23_gptfttest-worker-small-group-gptfttest-wxm29,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:logging_first_step=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:logging_nan_inf_filter=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:logging_steps=500,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:logging_strategy=steps,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:lr_scheduler_type=linear,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:max_grad_norm=1.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:max_steps=-1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:metric_for_best_model=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:mp_parameters=,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:no_cuda=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:num_train_epochs=3.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:optim=adamw_hf,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:optim_args=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:output_dir=/tmp/test-clm,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:overwrite_output_dir=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:past_index=-1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:per_device_eval_batch_size=2,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:per_device_train_batch_size=2,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:prediction_loss_only=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:push_to_hub=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:push_to_hub_model_id=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:push_to_hub_organization=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:push_to_hub_token=,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:ray_scope=last,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:remove_unused_columns=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:report_to=[\\'tensorboard\\'],\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:resume_from_checkpoint=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:run_name=/tmp/test-clm,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:save_on_each_node=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:save_safetensors=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:save_steps=500,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:save_strategy=steps,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:save_total_limit=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:seed=42,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:sharded_ddp=[],\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:skip_memory_metrics=True,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:tf32=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:torch_compile=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:torch_compile_backend=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:torch_compile_mode=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:torchdynamo=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:tpu_metrics_debug=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:tpu_num_cores=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:use_ipex=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:use_legacy_prediction_loop=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:use_mps_device=False,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:warmup_ratio=0.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:warmup_steps=0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:weight_decay=0.0,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:xpu_backend=None,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:)\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - https://huggingface.co/datasets/wikitext/resolve/main/wikitext.py not found in cache or force_download set to True, downloading to /home/ray/.cache/huggingface/datasets/downloads/tmpzmpwplcn\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - storing https://huggingface.co/datasets/wikitext/resolve/main/wikitext.py in cache at /home/ray/.cache/huggingface/datasets/downloads/30cb21e192e211952c02572882251280460fb5247fe18b6c0fb69224e769f1e1.6a998136b3179c543fac19963253d25970e7fe6d053f2818edc7075627f64bad.py\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - creating metadata file for /home/ray/.cache/huggingface/datasets/downloads/30cb21e192e211952c02572882251280460fb5247fe18b6c0fb69224e769f1e1.6a998136b3179c543fac19963253d25970e7fe6d053f2818edc7075627f64bad.py\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - https://huggingface.co/datasets/wikitext/resolve/main/dataset_infos.json not found in cache or force_download set to True, downloading to /home/ray/.cache/huggingface/datasets/downloads/tmpdadf8ql0\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - storing https://huggingface.co/datasets/wikitext/resolve/main/dataset_infos.json in cache at /home/ray/.cache/huggingface/datasets/downloads/87ea4775c52b60feb08a5087c68f4453d4533a02491172390b4d6a3f97ae44d1.d3aa47a864d0b5cf3b7ebcf51e45c9d8f96356ff8527fff02d3a4cae4c9f5b1e\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:04/26/2023 07:33:25 - INFO - datasets.utils.file_utils - creating metadata file for /home/ray/.cache/huggingface/datasets/downloads/87ea4775c52b60feb08a5087c68f4453d4533a02491172390b4d6a3f97ae44d1.d3aa47a864d0b5cf3b7ebcf51e45c9d8f96356ff8527fff02d3a4cae4c9f5b1e\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:ERROR: pip\\'s dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:googleapis-common-protos 1.57.0 requires protobuf!=3.20.0,!=3.20.1,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5, but you have protobuf 3.20.1 which is incompatible.\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:google-api-core 2.10.2 requires protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5, but you have protobuf 3.20.1 which is incompatible.\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:azure-cli-core 2.29.1 requires requests[socks]~=2.25.1, but you have requests 2.28.2 which is incompatible.\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[notice] A new release of pip available: 22.3.1 -> 23.1.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[notice] To update, run: pip install --upgrade pip\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Downloading builder script: 0%| | 0.00/8.48k [00:00> loading configuration file config.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:720] 2023-04-26 07:33:27,263 >> Model config GPT2Config {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"_name_or_path\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"activation_function\": \"gelu_new\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"architectures\": [\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"GPT2LMHeadModel\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ],\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"attn_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"embd_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"initializer_range\": 0.02,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"layer_norm_epsilon\": 1e-05,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"model_type\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_ctx\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_embd\": 768,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_head\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_inner\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_layer\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_positions\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"reorder_and_upcast_attn\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"resid_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"scale_attn_by_inverse_layer_idx\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"scale_attn_weights\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_activation\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_first_dropout\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_proj_to_labels\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_type\": \"cls_index\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_use_proj\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"task_specific_params\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"text-generation\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"do_sample\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"max_length\": 50\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: }\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: },\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"transformers_version\": \"4.28.1\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"use_cache\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"vocab_size\": 50257\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:}\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|tokenization_auto.py:502] 2023-04-26 07:33:27,297 >> Could not locate the tokenizer configuration file, will try to use the model config instead.\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:668] 2023-04-26 07:33:27,323 >> loading configuration file config.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:720] 2023-04-26 07:33:27,324 >> Model config GPT2Config {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"_name_or_path\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"activation_function\": \"gelu_new\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"architectures\": [\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"GPT2LMHeadModel\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ],\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"attn_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"embd_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"initializer_range\": 0.02,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"layer_norm_epsilon\": 1e-05,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"model_type\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_ctx\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_embd\": 768,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_head\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_inner\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_layer\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_positions\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"reorder_and_upcast_attn\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"resid_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"scale_attn_by_inverse_layer_idx\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"scale_attn_weights\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_activation\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_first_dropout\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_proj_to_labels\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_type\": \"cls_index\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_use_proj\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"task_specific_params\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"text-generation\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"do_sample\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"max_length\": 50\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: }\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: },\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"transformers_version\": \"4.28.1\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"use_cache\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"vocab_size\": 50257\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:}\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Downloading (…)olve/main/vocab.json: 0%| | 0.00/1.04M [00:00> loading file vocab.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/vocab.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:27,861 >> loading file merges.txt from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/merges.txt\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:27,861 >> loading file tokenizer.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/tokenizer.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:27,861 >> loading file added_tokens.json from cache at None\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:27,861 >> loading file special_tokens_map.json from cache at None\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:27,861 >> loading file tokenizer_config.json from cache at None\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:668] 2023-04-26 07:33:27,861 >> loading configuration file config.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:720] 2023-04-26 07:33:27,862 >> Model config GPT2Config {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"_name_or_path\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"activation_function\": \"gelu_new\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"architectures\": [\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"GPT2LMHeadModel\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: ],\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"attn_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"embd_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"initializer_range\": 0.02,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"layer_norm_epsilon\": 1e-05,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"model_type\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_ctx\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_embd\": 768,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_head\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_inner\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:ERROR: pip\\'s dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:googleapis-common-protos 1.57.0 requires protobuf!=3.20.0,!=3.20.1,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5, but you have protobuf 3.20.1 which is incompatible.\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:google-api-core 2.10.2 requires protobuf!=3.20.0,!=3.20.1,!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.19.5, but you have protobuf 3.20.1 which is incompatible.\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:azure-cli-core 2.29.1 requires requests[socks]~=2.25.1, but you have requests 2.28.2 which is incompatible.\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[notice] A new release of pip available: 22.3.1 -> 23.1.2\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[notice] To update, run: pip install --upgrade pip\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Downloading builder script: 0%| | 0.00/8.48k [00:00> loading configuration file config.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:720] 2023-04-26 07:33:27,429 >> Model config GPT2Config {\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"_name_or_path\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"activation_function\": \"gelu_new\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"architectures\": [\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"GPT2LMHeadModel\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ],\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"attn_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"embd_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"initializer_range\": 0.02,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"layer_norm_epsilon\": 1e-05,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"model_type\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_ctx\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_embd\": 768,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_head\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_inner\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_layer\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_positions\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"reorder_and_upcast_attn\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"resid_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"scale_attn_by_inverse_layer_idx\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"scale_attn_weights\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_activation\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_first_dropout\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_proj_to_labels\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_type\": \"cls_index\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_use_proj\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"task_specific_params\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"text-generation\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"do_sample\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"max_length\": 50\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: }\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: },\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"transformers_version\": \"4.28.1\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"use_cache\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"vocab_size\": 50257\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:}\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|tokenization_auto.py:502] 2023-04-26 07:33:27,475 >> Could not locate the tokenizer configuration file, will try to use the model config instead.\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:668] 2023-04-26 07:33:27,521 >> loading configuration file config.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:720] 2023-04-26 07:33:27,522 >> Model config GPT2Config {\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"_name_or_path\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"activation_function\": \"gelu_new\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"architectures\": [\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"GPT2LMHeadModel\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ],\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"attn_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"embd_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"initializer_range\": 0.02,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"layer_norm_epsilon\": 1e-05,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"model_type\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_ctx\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_embd\": 768,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_head\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_inner\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_layer\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_positions\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"reorder_and_upcast_attn\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"resid_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"scale_attn_by_inverse_layer_idx\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"scale_attn_weights\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_activation\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_first_dropout\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_proj_to_labels\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_type\": \"cls_index\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"summary_use_proj\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"task_specific_params\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"text-generation\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"do_sample\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"max_length\": 50\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: }\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: },\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"transformers_version\": \"4.28.1\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"use_cache\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"vocab_size\": 50257\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:}\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Downloading (…)olve/main/vocab.json: 0%| | 0.00/1.04M [00:00> loading file vocab.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/vocab.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:28,273 >> loading file merges.txt from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/merges.txt\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:28,273 >> loading file tokenizer.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/tokenizer.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:28,273 >> loading file added_tokens.json from cache at None\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:28,273 >> loading file special_tokens_map.json from cache at None\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:1809] 2023-04-26 07:33:28,273 >> loading file tokenizer_config.json from cache at None\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:668] 2023-04-26 07:33:28,273 >> loading configuration file config.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:720] 2023-04-26 07:33:28,274 >> Model config GPT2Config {\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"_name_or_path\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"activation_function\": \"gelu_new\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"architectures\": [\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"GPT2LMHeadModel\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: ],\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"attn_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"embd_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"initializer_range\": 0.02,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"layer_norm_epsilon\": 1e-05,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"model_type\": \"gpt2\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_ctx\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_embd\": 768,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_head\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_inner\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"n_layer\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_layer\": 12,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"n_positions\": 1024,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"reorder_and_upcast_attn\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"resid_pdrop\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"scale_attn_by_inverse_layer_idx\": false,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"scale_attn_weights\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_activation\": null,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_first_dropout\": 0.1,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_proj_to_labels\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_type\": \"cls_index\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"summary_use_proj\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"task_specific_params\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"text-generation\": {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"do_sample\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"max_length\": 50\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: }\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: },\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"transformers_version\": \"4.28.1\",\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"use_cache\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"vocab_size\": 50257\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:}\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Downloading pytorch_model.bin: 0%| | 0.00/548M [00:00> loading weights file pytorch_model.bin from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/pytorch_model.bin\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:575] 2023-04-26 07:33:31,079 >> Generate config GenerationConfig {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"_from_model_config\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"transformers_version\": \"4.28.1\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:}\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|modeling_utils.py:3190] 2023-04-26 07:33:32,989 >> All model checkpoint weights were used when initializing GPT2LMHeadModel.\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|modeling_utils.py:3198] 2023-04-26 07:33:32,989 >> All the weights of GPT2LMHeadModel were initialized from the model checkpoint at gpt2.\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:If your task is similar to the task the model of the checkpoint was trained on, you can already use GPT2LMHeadModel for predictions without further training.\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Downloading (…)neration_config.json: 0%| | 0.00/124 [00:00> loading configuration file generation_config.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/generation_config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:575] 2023-04-26 07:33:33,094 >> Generate config GenerationConfig {\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"_from_model_config\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \"transformers_version\": \"4.28.1\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:}\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:Running tokenizer on dataset: 0%| | 0/4358 [00:00> loading weights file pytorch_model.bin from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/pytorch_model.bin\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:575] 2023-04-26 07:33:31,686 >> Generate config GenerationConfig {\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"_from_model_config\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"transformers_version\": \"4.28.1\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:}\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|modeling_utils.py:3190] 2023-04-26 07:33:33,605 >> All model checkpoint weights were used when initializing GPT2LMHeadModel.\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|modeling_utils.py:3198] 2023-04-26 07:33:33,605 >> All the weights of GPT2LMHeadModel were initialized from the model checkpoint at gpt2.\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:If your task is similar to the task the model of the checkpoint was trained on, you can already use GPT2LMHeadModel for predictions without further training.\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Downloading (…)neration_config.json: 0%| | 0.00/124 [00:00> loading configuration file generation_config.json from cache at /home/ray/.cache/huggingface/hub/models--gpt2/snapshots/e7da7f221d5bf496a48136c0cd264e630fe9fcc8/generation_config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:575] 2023-04-26 07:33:33,714 >> Generate config GenerationConfig {\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"_from_model_config\": true,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"bos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"eos_token_id\": 50256,\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: \"transformers_version\": \"4.28.1\"\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:}\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:Running tokenizer on dataset: 0%| | 0/4358 [00:00> ***** Running training *****\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|trainer.py:1770] 2023-04-26 07:33:45,732 >> Num examples = 2,318\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|trainer.py:1771] 2023-04-26 07:33:45,732 >> Num Epochs = 3\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|trainer.py:1772] 2023-04-26 07:33:45,732 >> Instantaneous batch size per device = 2\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|trainer.py:1773] 2023-04-26 07:33:45,732 >> Total train batch size (w. parallel, distributed & accumulation) = 4\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|trainer.py:1774] 2023-04-26 07:33:45,733 >> Gradient Accumulation steps = 1\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|trainer.py:1775] 2023-04-26 07:33:45,733 >> Total optimization steps = 1,740\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|trainer.py:1776] 2023-04-26 07:33:45,733 >> Number of trainable parameters = 124,439,808\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 0%| | 0/1740 [00:00> ***** Running training *****\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|trainer.py:1770] 2023-04-26 07:33:45,729 >> Num examples = 2,318\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|trainer.py:1771] 2023-04-26 07:33:45,729 >> Num Epochs = 3\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|trainer.py:1772] 2023-04-26 07:33:45,729 >> Instantaneous batch size per device = 2\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|trainer.py:1773] 2023-04-26 07:33:45,729 >> Total train batch size (w. parallel, distributed & accumulation) = 4\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|trainer.py:1774] 2023-04-26 07:33:45,729 >> Gradient Accumulation steps = 1\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|trainer.py:1775] 2023-04-26 07:33:45,729 >> Total optimization steps = 1,740\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:[INFO|trainer.py:1776] 2023-04-26 07:33:45,729 >> Number of trainable parameters = 124,439,808\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 0%| | 0/1740 [00:00> Saving model checkpoint to /tmp/test-clm/checkpoint-500\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:457] 2023-04-26 07:42:32,569 >> Configuration saved in /tmp/test-clm/checkpoint-500/config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|configuration_utils.py:362] 2023-04-26 07:42:32,569 >> Configuration saved in /tmp/test-clm/checkpoint-500/generation_config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|modeling_utils.py:1847] 2023-04-26 07:42:33,212 >> Model weights saved in /tmp/test-clm/checkpoint-500/pytorch_model.bin\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:2171] 2023-04-26 07:42:33,212 >> tokenizer config file saved in /tmp/test-clm/checkpoint-500/tokenizer_config.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:[INFO|tokenization_utils_base.py:2178] 2023-04-26 07:42:33,213 >> Special tokens file saved in /tmp/test-clm/checkpoint-500/special_tokens_map.json\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 501/1740 [08:49<34:21, 1.66s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 502/1740 [08:50<30:28, 1.48s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 503/1740 [08:51<28:09, 1.37s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 504/1740 [08:52<26:28, 1.29s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 505/1740 [08:53<24:44, 1.20s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 506/1740 [08:54<23:23, 1.14s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 507/1740 [08:55<22:34, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 508/1740 [08:56<22:11, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 509/1740 [08:57<22:02, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 510/1740 [08:58<21:26, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 511/1740 [08:59<20:58, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 512/1740 [09:00<21:34, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 29%|██▉ | 513/1740 [09:01<21:46, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|██▉ | 514/1740 [09:02<21:45, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|██▉ | 515/1740 [09:03<20:52, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|██▉ | 516/1740 [09:04<20:44, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|██▉ | 517/1740 [09:05<21:02, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|██▉ | 518/1740 [09:06<21:18, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|██▉ | 519/1740 [09:08<21:09, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|██▉ | 520/1740 [09:09<20:54, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|██▉ | 521/1740 [09:09<20:34, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|███ | 522/1740 [09:11<21:54, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|███ | 523/1740 [09:12<21:27, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|███ | 524/1740 [09:13<21:37, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|███ | 525/1740 [09:14<21:46, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|███ | 526/1740 [09:15<22:20, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|███ | 527/1740 [09:16<21:23, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|███ | 528/1740 [09:17<21:43, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|███ | 529/1740 [09:18<21:18, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 30%|███ | 530/1740 [09:19<21:17, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 531/1740 [09:20<21:49, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 532/1740 [09:21<22:00, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 533/1740 [09:23<21:39, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 534/1740 [09:24<21:51, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 535/1740 [09:25<20:52, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 536/1740 [09:26<21:12, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 537/1740 [09:27<20:43, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▍ | 428/1740 [07:30<22:36, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▍ | 429/1740 [07:31<22:37, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▍ | 430/1740 [07:32<22:27, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▍ | 431/1740 [07:33<22:31, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▍ | 432/1740 [07:34<22:01, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▍ | 433/1740 [07:35<21:18, 1.02it/s][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▍ | 434/1740 [07:36<22:45, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▌ | 435/1740 [07:37<22:10, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▌ | 436/1740 [07:38<21:41, 1.00it/s][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▌ | 437/1740 [07:39<22:30, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▌ | 438/1740 [07:40<23:14, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▌ | 439/1740 [07:41<23:11, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▌ | 440/1740 [07:42<23:21, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▌ | 441/1740 [07:43<23:49, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▌ | 442/1740 [07:45<23:57, 1.11s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 25%|██▌ | 443/1740 [07:46<23:16, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 444/1740 [07:47<22:49, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 445/1740 [07:48<22:55, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 446/1740 [07:49<23:18, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 447/1740 [07:50<22:52, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 448/1740 [07:51<22:44, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 449/1740 [07:52<22:00, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 450/1740 [07:53<21:54, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 451/1740 [07:54<21:56, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 452/1740 [07:55<22:19, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 453/1740 [07:56<22:13, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 454/1740 [07:57<22:17, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 455/1740 [07:58<21:59, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▌ | 456/1740 [07:59<22:39, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▋ | 457/1740 [08:00<23:06, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▋ | 458/1740 [08:01<23:04, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▋ | 459/1740 [08:02<22:57, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▋ | 460/1740 [08:03<22:31, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 26%|██▋ | 461/1740 [08:05<23:03, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 462/1740 [08:06<22:42, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 463/1740 [08:06<21:54, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 464/1740 [08:08<22:04, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 465/1740 [08:09<22:21, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 466/1740 [08:10<21:42, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 467/1740 [08:11<22:28, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 468/1740 [08:12<22:26, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 469/1740 [08:13<22:33, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 470/1740 [08:14<22:17, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 471/1740 [08:15<23:02, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 472/1740 [08:16<22:43, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 473/1740 [08:17<22:32, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 474/1740 [08:18<22:41, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 475/1740 [08:19<22:07, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 476/1740 [08:20<22:37, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 477/1740 [08:21<22:00, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 27%|██▋ | 478/1740 [08:22<21:31, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 479/1740 [08:23<22:07, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 480/1740 [08:24<21:14, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 481/1740 [08:25<21:50, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 482/1740 [08:27<22:07, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 483/1740 [08:28<22:34, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 484/1740 [08:29<21:50, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 485/1740 [08:30<22:05, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 486/1740 [08:31<22:01, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 487/1740 [08:32<21:35, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 488/1740 [08:33<21:47, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 489/1740 [08:34<21:43, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 490/1740 [08:35<21:50, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 491/1740 [08:36<22:43, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 492/1740 [08:37<21:52, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 493/1740 [08:38<22:22, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 494/1740 [08:39<21:58, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 28%|██▊ | 495/1740 [08:40<22:40, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▊ | 496/1740 [08:42<22:56, 1.11s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▊ | 497/1740 [08:43<23:04, 1.11s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▊ | 498/1740 [08:44<22:12, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▊ | 499/1740 [08:45<22:39, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▊ | 500/1740 [08:46<21:57, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: \\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▊ | 500/1740 [08:46<21:57, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 501/1740 [08:49<34:20, 1.66s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 502/1740 [08:50<30:28, 1.48s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 503/1740 [08:51<28:09, 1.37s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 504/1740 [08:52<26:28, 1.29s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 505/1740 [08:53<24:44, 1.20s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 506/1740 [08:54<23:24, 1.14s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 507/1740 [08:55<22:34, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 508/1740 [08:56<22:11, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 509/1740 [08:57<22:03, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 510/1740 [08:58<21:26, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 511/1740 [08:59<20:58, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 512/1740 [09:00<21:34, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 29%|██▉ | 513/1740 [09:01<21:45, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|██▉ | 514/1740 [09:02<21:45, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|██▉ | 515/1740 [09:03<20:53, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|██▉ | 516/1740 [09:04<20:46, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|██▉ | 517/1740 [09:05<21:01, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|██▉ | 518/1740 [09:06<21:17, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|██▉ | 519/1740 [09:08<21:08, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|██▉ | 520/1740 [09:09<20:54, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|██▉ | 521/1740 [09:10<20:33, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|███ | 522/1740 [09:11<21:54, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|███ | 523/1740 [09:12<21:27, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|███ | 524/1740 [09:13<21:37, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|███ | 525/1740 [09:14<21:46, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|███ | 526/1740 [09:15<22:20, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|███ | 527/1740 [09:16<21:23, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|███ | 528/1740 [09:17<21:43, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|███ | 529/1740 [09:18<21:18, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 30%|███ | 530/1740 [09:19<21:17, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 531/1740 [09:20<21:49, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 532/1740 [09:21<22:00, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 533/1740 [09:23<21:39, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 534/1740 [09:24<21:51, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 535/1740 [09:25<20:52, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 536/1740 [09:26<21:12, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 537/1740 [09:27<20:43, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 538/1740 [09:28<20:31, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 539/1740 [09:29<20:37, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 540/1740 [09:30<20:25, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 541/1740 [09:31<20:48, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 542/1740 [09:32<21:04, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███ | 543/1740 [09:33<21:06, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███▏ | 544/1740 [09:34<21:21, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███▏ | 545/1740 [09:35<21:15, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███▏ | 546/1740 [09:36<20:38, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███▏ | 547/1740 [09:37<20:24, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 31%|███▏ | 548/1740 [09:38<20:28, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 549/1740 [09:39<21:00, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 550/1740 [09:40<21:25, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 538/1740 [09:28<20:31, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 539/1740 [09:29<20:37, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 540/1740 [09:30<20:25, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 541/1740 [09:31<20:47, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 542/1740 [09:32<21:05, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███ | 543/1740 [09:33<21:06, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███▏ | 544/1740 [09:34<21:21, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███▏ | 545/1740 [09:35<21:15, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███▏ | 546/1740 [09:36<20:39, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███▏ | 547/1740 [09:37<20:24, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 31%|███▏ | 548/1740 [09:38<20:28, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 549/1740 [09:39<21:01, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 550/1740 [09:40<21:25, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 551/1740 [09:41<20:49, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 552/1740 [09:42<20:14, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 553/1740 [09:43<20:22, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 554/1740 [09:44<20:14, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 555/1740 [09:45<20:15, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 556/1740 [09:46<20:07, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 557/1740 [09:47<20:31, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 558/1740 [09:49<20:25, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 559/1740 [09:50<20:24, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 560/1740 [09:51<20:54, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 561/1740 [09:52<20:39, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 562/1740 [09:53<20:55, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 563/1740 [09:54<20:42, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 564/1740 [09:55<19:57, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 32%|███▏ | 565/1740 [09:56<19:49, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 566/1740 [09:57<19:47, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 567/1740 [09:58<20:11, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 568/1740 [09:59<19:56, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 569/1740 [10:00<20:27, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 570/1740 [10:01<20:51, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 571/1740 [10:02<20:51, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 572/1740 [10:03<20:22, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 573/1740 [10:04<20:14, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 574/1740 [10:05<20:35, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 575/1740 [10:06<20:37, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 576/1740 [10:07<19:49, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 577/1740 [10:08<20:27, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 578/1740 [10:09<19:58, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 579/1740 [10:10<20:00, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 580/1740 [10:11<19:31, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 581/1740 [10:12<19:45, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 33%|███▎ | 582/1740 [10:13<19:49, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▎ | 583/1740 [10:14<19:52, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▎ | 584/1740 [10:16<19:53, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▎ | 585/1740 [10:17<19:41, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▎ | 586/1740 [10:18<19:46, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▎ | 587/1740 [10:19<19:51, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 588/1740 [10:20<19:29, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 589/1740 [10:21<19:28, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 590/1740 [10:22<19:25, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 591/1740 [10:23<20:17, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 592/1740 [10:24<20:16, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 593/1740 [10:25<19:42, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 594/1740 [10:26<19:50, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 595/1740 [10:27<20:15, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 596/1740 [10:28<20:23, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 597/1740 [10:29<20:42, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 598/1740 [10:30<20:43, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 599/1740 [10:31<20:27, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 34%|███▍ | 600/1740 [10:32<20:29, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▍ | 601/1740 [10:33<19:57, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▍ | 602/1740 [10:34<20:03, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▍ | 603/1740 [10:35<19:48, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▍ | 604/1740 [10:37<20:31, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▍ | 605/1740 [10:38<20:26, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▍ | 606/1740 [10:39<19:52, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▍ | 607/1740 [10:40<19:38, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▍ | 608/1740 [10:41<19:47, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▌ | 609/1740 [10:42<19:44, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▌ | 610/1740 [10:43<20:12, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▌ | 611/1740 [10:44<19:41, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▌ | 612/1740 [10:45<20:09, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▌ | 613/1740 [10:46<20:06, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▌ | 614/1740 [10:47<19:27, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▌ | 615/1740 [10:48<19:31, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▌ | 616/1740 [10:49<19:44, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 35%|███▌ | 617/1740 [10:50<20:07, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 618/1740 [10:52<20:35, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 619/1740 [10:53<20:06, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 620/1740 [10:54<19:53, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 621/1740 [10:55<19:27, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 622/1740 [10:56<19:55, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 623/1740 [10:57<19:45, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 624/1740 [10:58<20:03, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 625/1740 [10:59<19:24, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 626/1740 [11:00<19:35, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 627/1740 [11:01<20:01, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 628/1740 [11:02<20:08, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 629/1740 [11:03<19:50, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▌ | 630/1740 [11:04<19:21, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▋ | 631/1740 [11:05<19:36, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▋ | 632/1740 [11:06<19:20, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▋ | 633/1740 [11:07<19:02, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▋ | 634/1740 [11:08<18:52, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 36%|███▋ | 635/1740 [11:09<19:06, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 636/1740 [11:10<19:13, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 637/1740 [11:12<21:09, 1.15s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 638/1740 [11:13<20:42, 1.13s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 639/1740 [11:14<20:45, 1.13s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 640/1740 [11:15<20:53, 1.14s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 641/1740 [11:16<20:26, 1.12s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 642/1740 [11:17<20:07, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 643/1740 [11:18<19:36, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 644/1740 [11:19<19:32, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 645/1740 [11:20<19:28, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 646/1740 [11:22<20:09, 1.11s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 647/1740 [11:23<19:42, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 648/1740 [11:24<19:23, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 649/1740 [11:25<19:24, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 650/1740 [11:26<19:31, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 651/1740 [11:27<19:02, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 37%|███▋ | 652/1740 [11:28<19:45, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 653/1740 [11:29<19:22, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 654/1740 [11:30<19:16, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 655/1740 [11:31<19:44, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 656/1740 [11:32<19:10, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 657/1740 [11:33<19:10, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 658/1740 [11:34<19:04, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 659/1740 [11:35<18:34, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 551/1740 [09:41<20:50, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 552/1740 [09:42<20:13, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 553/1740 [09:43<20:22, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 554/1740 [09:44<20:14, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 555/1740 [09:45<20:14, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 556/1740 [09:46<20:07, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 557/1740 [09:47<20:31, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 558/1740 [09:49<20:25, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 559/1740 [09:50<20:24, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 560/1740 [09:51<20:54, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 561/1740 [09:52<20:39, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 562/1740 [09:53<20:55, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 563/1740 [09:54<20:42, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 564/1740 [09:55<19:56, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 32%|███▏ | 565/1740 [09:56<19:50, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 566/1740 [09:57<19:50, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 567/1740 [09:58<20:11, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 568/1740 [09:59<19:55, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 569/1740 [10:00<20:28, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 570/1740 [10:01<20:50, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 571/1740 [10:02<20:51, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 572/1740 [10:03<20:22, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 573/1740 [10:04<20:14, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 574/1740 [10:05<20:35, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 575/1740 [10:06<20:36, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 576/1740 [10:07<19:49, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 577/1740 [10:08<20:27, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 578/1740 [10:09<19:58, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 579/1740 [10:10<20:00, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 580/1740 [10:11<19:31, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 581/1740 [10:12<19:45, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 33%|███▎ | 582/1740 [10:13<19:49, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▎ | 583/1740 [10:14<19:52, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▎ | 584/1740 [10:16<19:53, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▎ | 585/1740 [10:17<19:41, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▎ | 586/1740 [10:18<19:46, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▎ | 587/1740 [10:19<19:51, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 588/1740 [10:20<19:29, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 589/1740 [10:21<19:28, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 590/1740 [10:22<19:26, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 591/1740 [10:23<20:18, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 592/1740 [10:24<20:15, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 593/1740 [10:25<19:43, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 594/1740 [10:26<19:50, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 595/1740 [10:27<20:15, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 596/1740 [10:28<20:23, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 597/1740 [10:29<20:43, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 598/1740 [10:30<20:43, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 599/1740 [10:31<20:27, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 34%|███▍ | 600/1740 [10:32<20:29, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▍ | 601/1740 [10:33<19:56, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▍ | 602/1740 [10:34<20:03, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▍ | 603/1740 [10:35<19:48, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▍ | 604/1740 [10:37<20:31, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▍ | 605/1740 [10:38<20:26, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▍ | 606/1740 [10:39<19:52, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▍ | 607/1740 [10:40<19:39, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▍ | 608/1740 [10:41<19:47, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▌ | 609/1740 [10:42<19:44, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▌ | 610/1740 [10:43<20:12, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▌ | 611/1740 [10:44<19:41, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▌ | 612/1740 [10:45<20:09, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▌ | 613/1740 [10:46<20:06, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▌ | 614/1740 [10:47<19:27, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▌ | 615/1740 [10:48<19:30, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▌ | 616/1740 [10:49<19:44, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 35%|███▌ | 617/1740 [10:50<20:07, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 618/1740 [10:52<20:34, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 619/1740 [10:53<20:05, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 620/1740 [10:54<19:53, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 621/1740 [10:55<19:27, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 622/1740 [10:56<19:54, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 623/1740 [10:57<19:45, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 624/1740 [10:58<20:03, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 625/1740 [10:59<19:24, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 626/1740 [11:00<19:35, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 627/1740 [11:01<20:01, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 628/1740 [11:02<20:08, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 629/1740 [11:03<19:50, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▌ | 630/1740 [11:04<19:21, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▋ | 631/1740 [11:05<19:35, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▋ | 632/1740 [11:06<19:20, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▋ | 633/1740 [11:07<19:02, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▋ | 634/1740 [11:08<18:52, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 36%|███▋ | 635/1740 [11:09<19:06, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 636/1740 [11:10<19:14, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 637/1740 [11:12<21:09, 1.15s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 638/1740 [11:13<20:42, 1.13s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 639/1740 [11:14<20:45, 1.13s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 640/1740 [11:15<20:52, 1.14s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 641/1740 [11:16<20:27, 1.12s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 642/1740 [11:17<20:07, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 643/1740 [11:18<19:36, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 644/1740 [11:19<19:33, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 645/1740 [11:20<19:28, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 646/1740 [11:22<20:08, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 647/1740 [11:23<19:42, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 648/1740 [11:24<19:23, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 649/1740 [11:25<19:24, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 650/1740 [11:26<19:31, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 651/1740 [11:27<19:02, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 37%|███▋ | 652/1740 [11:28<19:45, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 653/1740 [11:29<19:23, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 654/1740 [11:30<19:16, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 655/1740 [11:31<19:44, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 656/1740 [11:32<19:10, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 657/1740 [11:33<19:10, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 658/1740 [11:34<19:04, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 659/1740 [11:35<18:34, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 660/1740 [11:36<18:56, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 661/1740 [11:37<18:49, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 662/1740 [11:39<19:04, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 663/1740 [11:40<19:03, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 664/1740 [11:41<19:18, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 665/1740 [11:42<19:23, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 666/1740 [11:43<19:27, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 667/1740 [11:44<19:21, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 668/1740 [11:45<18:55, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 38%|███▊ | 669/1740 [11:46<18:18, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▊ | 670/1740 [11:47<18:06, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▊ | 671/1740 [11:48<18:18, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▊ | 672/1740 [11:49<18:30, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 660/1740 [11:36<18:56, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 661/1740 [11:37<18:49, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 662/1740 [11:39<19:04, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 663/1740 [11:40<19:03, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 664/1740 [11:41<19:18, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 665/1740 [11:42<19:23, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 666/1740 [11:43<19:28, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 667/1740 [11:44<19:21, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 668/1740 [11:45<18:55, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 38%|███▊ | 669/1740 [11:46<18:18, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▊ | 670/1740 [11:47<18:06, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▊ | 671/1740 [11:48<18:18, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▊ | 672/1740 [11:49<18:29, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▊ | 673/1740 [11:50<18:48, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▊ | 674/1740 [11:51<18:20, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 675/1740 [11:52<18:18, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 676/1740 [11:53<18:39, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 677/1740 [11:54<18:04, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 678/1740 [11:55<18:14, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 679/1740 [11:56<17:43, 1.00s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 680/1740 [11:57<17:33, 1.01it/s][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 681/1740 [11:58<18:07, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 682/1740 [11:59<18:01, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 683/1740 [12:00<18:23, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 684/1740 [12:01<18:24, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 685/1740 [12:03<18:54, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 686/1740 [12:04<18:10, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 39%|███▉ | 687/1740 [12:05<18:04, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|███▉ | 688/1740 [12:06<18:33, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|███▉ | 689/1740 [12:07<18:23, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|███▉ | 690/1740 [12:08<18:09, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|███▉ | 691/1740 [12:09<17:40, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|███▉ | 692/1740 [12:10<17:22, 1.01it/s][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|███▉ | 693/1740 [12:11<17:33, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|███▉ | 694/1740 [12:12<18:20, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|███▉ | 695/1740 [12:13<18:50, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|████ | 696/1740 [12:14<18:51, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|████ | 697/1740 [12:15<18:55, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|████ | 698/1740 [12:16<18:43, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|████ | 699/1740 [12:17<18:31, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|████ | 700/1740 [12:18<18:25, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|████ | 701/1740 [12:19<18:15, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|████ | 702/1740 [12:20<18:04, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|████ | 703/1740 [12:22<18:42, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 40%|████ | 704/1740 [12:23<19:19, 1.12s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 705/1740 [12:24<19:19, 1.12s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 706/1740 [12:25<19:01, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 707/1740 [12:26<18:37, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 708/1740 [12:27<18:33, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 709/1740 [12:28<17:57, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 710/1740 [12:29<17:50, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 711/1740 [12:30<17:36, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 712/1740 [12:31<18:15, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 713/1740 [12:32<18:33, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 714/1740 [12:33<18:35, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 715/1740 [12:34<17:55, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 716/1740 [12:35<18:08, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████ | 717/1740 [12:36<17:55, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████▏ | 718/1740 [12:38<18:32, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████▏ | 719/1740 [12:39<18:04, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████▏ | 720/1740 [12:40<17:42, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████▏ | 721/1740 [12:41<17:51, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 41%|████▏ | 722/1740 [12:42<17:32, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 723/1740 [12:43<17:52, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 724/1740 [12:44<18:00, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 725/1740 [12:45<18:08, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 726/1740 [12:46<18:14, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 727/1740 [12:47<17:59, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 728/1740 [12:48<17:49, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 729/1740 [12:49<17:40, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 730/1740 [12:50<17:41, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 731/1740 [12:51<17:24, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 732/1740 [12:52<17:53, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 733/1740 [12:53<17:38, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 734/1740 [12:55<17:53, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 735/1740 [12:56<18:06, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 736/1740 [12:57<17:56, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 737/1740 [12:58<18:00, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 738/1740 [12:59<17:12, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 42%|████▏ | 739/1740 [13:00<18:09, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 740/1740 [13:01<18:08, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 741/1740 [13:02<18:08, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 742/1740 [13:03<18:05, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 743/1740 [13:04<17:51, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 744/1740 [13:05<17:13, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 745/1740 [13:06<17:20, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 746/1740 [13:07<17:10, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 747/1740 [13:08<17:39, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 748/1740 [13:09<17:29, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 749/1740 [13:10<17:18, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 750/1740 [13:12<17:42, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 751/1740 [13:13<17:52, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 752/1740 [13:14<17:49, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 753/1740 [13:15<18:02, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 754/1740 [13:16<18:08, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 755/1740 [13:17<17:18, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 43%|████▎ | 756/1740 [13:18<17:02, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▎ | 757/1740 [13:19<16:52, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▎ | 758/1740 [13:20<16:38, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▎ | 759/1740 [13:21<16:41, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▎ | 760/1740 [13:22<17:01, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▎ | 761/1740 [13:23<18:16, 1.12s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 762/1740 [13:24<18:06, 1.11s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 763/1740 [13:26<18:01, 1.11s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 764/1740 [13:27<17:52, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 765/1740 [13:28<18:04, 1.11s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 766/1740 [13:29<17:20, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 767/1740 [13:30<17:23, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 768/1740 [13:31<17:22, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 769/1740 [13:32<17:13, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 770/1740 [13:33<17:21, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 771/1740 [13:34<17:06, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 772/1740 [13:35<16:53, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 773/1740 [13:36<17:01, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 44%|████▍ | 774/1740 [13:37<16:40, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 45%|████▍ | 775/1740 [13:38<17:34, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 45%|████▍ | 776/1740 [13:39<17:13, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 45%|████▍ | 777/1740 [13:41<17:19, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 45%|████▍ | 778/1740 [13:42<17:26, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=120, ip=10.129.22.18)\\x1b[0m [0]: 45%|████▍ | 779/1740 [13:43<17:12, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▊ | 673/1740 [11:50<18:49, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▊ | 674/1740 [11:51<18:20, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 675/1740 [11:52<18:18, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 676/1740 [11:53<18:39, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 677/1740 [11:54<18:04, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 678/1740 [11:55<18:14, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 679/1740 [11:56<17:42, 1.00s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 680/1740 [11:57<17:33, 1.01it/s][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 681/1740 [11:58<18:07, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 682/1740 [11:59<18:00, 1.02s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 683/1740 [12:00<18:23, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 684/1740 [12:01<18:24, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 685/1740 [12:03<18:54, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 686/1740 [12:04<18:11, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 39%|███▉ | 687/1740 [12:05<18:03, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|███▉ | 688/1740 [12:06<18:33, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|███▉ | 689/1740 [12:07<18:23, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|███▉ | 690/1740 [12:08<18:10, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|███▉ | 691/1740 [12:09<17:40, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|███▉ | 692/1740 [12:10<17:22, 1.01it/s][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|███▉ | 693/1740 [12:11<17:33, 1.01s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|███▉ | 694/1740 [12:12<18:20, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|███▉ | 695/1740 [12:13<18:52, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|████ | 696/1740 [12:14<18:50, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|████ | 697/1740 [12:15<18:55, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|████ | 698/1740 [12:16<18:46, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|████ | 699/1740 [12:17<18:32, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|████ | 700/1740 [12:18<18:24, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|████ | 701/1740 [12:19<18:15, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|████ | 702/1740 [12:20<18:04, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|████ | 703/1740 [12:22<18:41, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 40%|████ | 704/1740 [12:23<19:19, 1.12s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 705/1740 [12:24<19:19, 1.12s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 706/1740 [12:25<19:00, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 707/1740 [12:26<18:37, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 708/1740 [12:27<18:33, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 709/1740 [12:28<17:57, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 710/1740 [12:29<17:50, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 711/1740 [12:30<17:35, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 712/1740 [12:31<18:15, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 713/1740 [12:32<18:33, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 714/1740 [12:33<18:35, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 715/1740 [12:34<17:55, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 716/1740 [12:35<18:08, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████ | 717/1740 [12:36<17:55, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████▏ | 718/1740 [12:38<18:32, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████▏ | 719/1740 [12:39<18:04, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████▏ | 720/1740 [12:40<17:42, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████▏ | 721/1740 [12:41<17:52, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 41%|████▏ | 722/1740 [12:42<17:31, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 723/1740 [12:43<17:52, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 724/1740 [12:44<18:00, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 725/1740 [12:45<18:08, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 726/1740 [12:46<18:14, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 727/1740 [12:47<18:00, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 728/1740 [12:48<17:49, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 729/1740 [12:49<17:40, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 730/1740 [12:50<17:41, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 731/1740 [12:51<17:23, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 732/1740 [12:52<17:53, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 733/1740 [12:53<17:38, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 734/1740 [12:55<17:54, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 735/1740 [12:56<18:06, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 736/1740 [12:57<17:57, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 737/1740 [12:58<18:00, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 738/1740 [12:59<17:11, 1.03s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 42%|████▏ | 739/1740 [13:00<18:09, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 740/1740 [13:01<18:08, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 741/1740 [13:02<18:08, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 742/1740 [13:03<18:05, 1.09s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 743/1740 [13:04<17:50, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 744/1740 [13:05<17:13, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 745/1740 [13:06<17:20, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 746/1740 [13:07<17:10, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 747/1740 [13:08<17:40, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 748/1740 [13:09<17:29, 1.06s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 749/1740 [13:10<17:18, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 750/1740 [13:12<17:41, 1.07s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 751/1740 [13:13<17:51, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 752/1740 [13:14<17:49, 1.08s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 753/1740 [13:15<18:02, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 754/1740 [13:16<18:08, 1.10s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 755/1740 [13:17<17:18, 1.05s/it][0]:\\n\\x1b[2m\\x1b[36m(CommandActor pid=121, ip=10.128.22.18)\\x1b[0m [0]: 43%|████▎ | 756/1740 [13:18<17:02, 1.04s/it][0]:\\n\\x1b[2m\\x1b[36m\n", + " -- TRUNCATED FOR FILE SIZE" + ] + }, + "execution_count": 674, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "job.logs()" + ] + }, + { + "cell_type": "code", + "execution_count": 658, + "id": "beb1a6b9-d9b3-49b7-b036-09f1d3569b59", + "metadata": {}, + "outputs": [], + "source": [ + "cluster.down()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "8398d977-db24-46d0-a7d2-b4e9197808d7", + "metadata": {}, + "outputs": [], + "source": [ + "auth.logout()" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.13" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/gpt_og.py b/demo-notebooks/guided-demos/notebook-ex-outputs/gpt_og.py new file mode 100644 index 000000000..d69e41fcb --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/gpt_og.py @@ -0,0 +1,728 @@ +#!/usr/bin/env python +# coding=utf-8 +# Copyright 2020 The HuggingFace Inc. team. All rights reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +""" +Fine-tuning the library models for causal language modeling (GPT, GPT-2, CTRL, ...) on a text file or a dataset. + +Here is the full list of checkpoints on the hub that can be fine-tuned by this script: +https://huggingface.co/models?filter=text-generation +""" +# You can also adapt this script on your own causal language modeling task. Pointers for this are left as comments. + +import subprocess + +subprocess.run(["pip", "uninstall", "protobuf"]) +subprocess.run( + [ + "pip", + "install", + "--upgrade", + "--target=/home/ray/workspace", + "-r", + "requirements.txt", + ] +) + +import logging +import math +import os +import sys +from dataclasses import dataclass, field +from itertools import chain +from typing import Optional + +import datasets +import evaluate +import torch +from datasets import load_dataset + +import transformers +from transformers import ( + CONFIG_MAPPING, + MODEL_FOR_CAUSAL_LM_MAPPING, + AutoConfig, + AutoModelForCausalLM, + AutoTokenizer, + HfArgumentParser, + Trainer, + TrainingArguments, + default_data_collator, + is_torch_tpu_available, + set_seed, +) +from transformers.testing_utils import CaptureLogger +from transformers.trainer_utils import get_last_checkpoint +from transformers.utils import check_min_version, send_example_telemetry +from transformers.utils.versions import require_version + + +# Will error if the minimal version of Transformers is not installed. Remove at your own risks. +# check_min_version("4.29.0.dev0") + +require_version( + "datasets>=1.8.0", + "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt", +) + +logger = logging.getLogger(__name__) + + +MODEL_CONFIG_CLASSES = list(MODEL_FOR_CAUSAL_LM_MAPPING.keys()) +MODEL_TYPES = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES) + + +@dataclass +class ModelArguments: + """ + Arguments pertaining to which model/config/tokenizer we are going to fine-tune, or train from scratch. + """ + + model_name_or_path: Optional[str] = field( + default=None, + metadata={ + "help": ( + "The model checkpoint for weights initialization.Don't set if you want to train a model from scratch." + ) + }, + ) + model_type: Optional[str] = field( + default=None, + metadata={ + "help": "If training from scratch, pass a model type from the list: " + + ", ".join(MODEL_TYPES) + }, + ) + config_overrides: Optional[str] = field( + default=None, + metadata={ + "help": ( + "Override some existing default config settings when a model is trained from scratch. Example: " + "n_embd=10,resid_pdrop=0.2,scale_attn_weights=false,summary_type=cls_index" + ) + }, + ) + config_name: Optional[str] = field( + default=None, + metadata={ + "help": "Pretrained config name or path if not the same as model_name" + }, + ) + tokenizer_name: Optional[str] = field( + default=None, + metadata={ + "help": "Pretrained tokenizer name or path if not the same as model_name" + }, + ) + cache_dir: Optional[str] = field( + default=None, + metadata={ + "help": "Where do you want to store the pretrained models downloaded from huggingface.co" + }, + ) + use_fast_tokenizer: bool = field( + default=True, + metadata={ + "help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not." + }, + ) + model_revision: str = field( + default="main", + metadata={ + "help": "The specific model version to use (can be a branch name, tag name or commit id)." + }, + ) + use_auth_token: bool = field( + default=False, + metadata={ + "help": ( + "Will use the token generated when running `huggingface-cli login` (necessary to use this script " + "with private models)." + ) + }, + ) + torch_dtype: Optional[str] = field( + default=None, + metadata={ + "help": ( + "Override the default `torch.dtype` and load the model under this dtype. If `auto` is passed, the " + "dtype will be automatically derived from the model's weights." + ), + "choices": ["auto", "bfloat16", "float16", "float32"], + }, + ) + low_cpu_mem_usage: bool = field( + default=False, + metadata={ + "help": ( + "It is an option to create the model as an empty shell, then only materialize its parameters when the pretrained weights are loaded." + "set True will benefit LLM loading time and RAM consumption." + ) + }, + ) + + def __post_init__(self): + if self.config_overrides is not None and ( + self.config_name is not None or self.model_name_or_path is not None + ): + raise ValueError( + "--config_overrides can't be used in combination with --config_name or --model_name_or_path" + ) + + +@dataclass +class DataTrainingArguments: + """ + Arguments pertaining to what data we are going to input our model for training and eval. + """ + + dataset_name: Optional[str] = field( + default=None, + metadata={"help": "The name of the dataset to use (via the datasets library)."}, + ) + dataset_config_name: Optional[str] = field( + default=None, + metadata={ + "help": "The configuration name of the dataset to use (via the datasets library)." + }, + ) + train_file: Optional[str] = field( + default=None, metadata={"help": "The input training data file (a text file)."} + ) + validation_file: Optional[str] = field( + default=None, + metadata={ + "help": "An optional input evaluation data file to evaluate the perplexity on (a text file)." + }, + ) + max_train_samples: Optional[int] = field( + default=None, + metadata={ + "help": ( + "For debugging purposes or quicker training, truncate the number of training examples to this " + "value if set." + ) + }, + ) + max_eval_samples: Optional[int] = field( + default=None, + metadata={ + "help": ( + "For debugging purposes or quicker training, truncate the number of evaluation examples to this " + "value if set." + ) + }, + ) + streaming: bool = field(default=False, metadata={"help": "Enable streaming mode"}) + block_size: Optional[int] = field( + default=None, + metadata={ + "help": ( + "Optional input sequence length after tokenization. " + "The training dataset will be truncated in block of this size for training. " + "Default to the model max input length for single sentence inputs (take into account special tokens)." + ) + }, + ) + overwrite_cache: bool = field( + default=False, + metadata={"help": "Overwrite the cached training and evaluation sets"}, + ) + validation_split_percentage: Optional[int] = field( + default=5, + metadata={ + "help": "The percentage of the train set used as validation set in case there's no validation split" + }, + ) + preprocessing_num_workers: Optional[int] = field( + default=None, + metadata={"help": "The number of processes to use for the preprocessing."}, + ) + keep_linebreaks: bool = field( + default=True, + metadata={"help": "Whether to keep line breaks when using TXT files or not."}, + ) + + def __post_init__(self): + if self.streaming: + require_version( + "datasets>=2.0.0", "The streaming feature requires `datasets>=2.0.0`" + ) + + if ( + self.dataset_name is None + and self.train_file is None + and self.validation_file is None + ): + raise ValueError( + "Need either a dataset name or a training/validation file." + ) + else: + if self.train_file is not None: + extension = self.train_file.split(".")[-1] + assert extension in [ + "csv", + "json", + "txt", + ], "`train_file` should be a csv, a json or a txt file." + if self.validation_file is not None: + extension = self.validation_file.split(".")[-1] + assert extension in [ + "csv", + "json", + "txt", + ], "`validation_file` should be a csv, a json or a txt file." + + +def main(): + # See all possible arguments in src/transformers/training_args.py + # or by passing the --help flag to this script. + # We now keep distinct sets of args, for a cleaner separation of concerns. + + parser = HfArgumentParser( + (ModelArguments, DataTrainingArguments, TrainingArguments) + ) + if len(sys.argv) == 2 and sys.argv[1].endswith(".json"): + # If we pass only one argument to the script and it's the path to a json file, + # let's parse it to get our arguments. + model_args, data_args, training_args = parser.parse_json_file( + json_file=os.path.abspath(sys.argv[1]) + ) + else: + model_args, data_args, training_args = parser.parse_args_into_dataclasses() + + # Sending telemetry. Tracking the example usage helps us better allocate resources to maintain them. The + # information sent is the one passed as arguments along with your Python/PyTorch versions. + send_example_telemetry("run_clm", model_args, data_args) + + # Setup logging + logging.basicConfig( + format="%(asctime)s - %(levelname)s - %(name)s - %(message)s", + datefmt="%m/%d/%Y %H:%M:%S", + handlers=[logging.StreamHandler(sys.stdout)], + ) + + if training_args.should_log: + # The default of training_args.log_level is passive, so we set log level at info here to have that default. + transformers.utils.logging.set_verbosity_info() + + log_level = training_args.get_process_log_level() + logger.setLevel(log_level) + datasets.utils.logging.set_verbosity(log_level) + transformers.utils.logging.set_verbosity(log_level) + transformers.utils.logging.enable_default_handler() + transformers.utils.logging.enable_explicit_format() + + # Log on each process the small summary: + logger.warning( + f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}" + + f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}" + ) + logger.info(f"Training/evaluation parameters {training_args}") + + # Detecting last checkpoint. + last_checkpoint = None + if ( + os.path.isdir(training_args.output_dir) + and training_args.do_train + and not training_args.overwrite_output_dir + ): + last_checkpoint = get_last_checkpoint(training_args.output_dir) + if last_checkpoint is None and len(os.listdir(training_args.output_dir)) > 0: + raise ValueError( + f"Output directory ({training_args.output_dir}) already exists and is not empty. " + "Use --overwrite_output_dir to overcome." + ) + elif ( + last_checkpoint is not None and training_args.resume_from_checkpoint is None + ): + logger.info( + f"Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change " + "the `--output_dir` or add `--overwrite_output_dir` to train from scratch." + ) + + # Set seed before initializing model. + set_seed(training_args.seed) + + # Get the datasets: you can either provide your own CSV/JSON/TXT training and evaluation files (see below) + # or just provide the name of one of the public datasets available on the hub at https://huggingface.co/datasets/ + # (the dataset will be downloaded automatically from the datasets Hub). + # + # For CSV/JSON files, this script will use the column called 'text' or the first column if no column called + # 'text' is found. You can easily tweak this behavior (see below). + # + # In distributed training, the load_dataset function guarantee that only one local process can concurrently + # download the dataset. + if data_args.dataset_name is not None: + # Downloading and loading a dataset from the hub. + raw_datasets = load_dataset( + data_args.dataset_name, + data_args.dataset_config_name, + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + streaming=data_args.streaming, + ) + if "validation" not in raw_datasets.keys(): + raw_datasets["validation"] = load_dataset( + data_args.dataset_name, + data_args.dataset_config_name, + split=f"train[:{data_args.validation_split_percentage}%]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + streaming=data_args.streaming, + ) + raw_datasets["train"] = load_dataset( + data_args.dataset_name, + data_args.dataset_config_name, + split=f"train[{data_args.validation_split_percentage}%:]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + streaming=data_args.streaming, + ) + else: + data_files = {} + dataset_args = {} + if data_args.train_file is not None: + data_files["train"] = data_args.train_file + if data_args.validation_file is not None: + data_files["validation"] = data_args.validation_file + extension = ( + data_args.train_file.split(".")[-1] + if data_args.train_file is not None + else data_args.validation_file.split(".")[-1] + ) + if extension == "txt": + extension = "text" + dataset_args["keep_linebreaks"] = data_args.keep_linebreaks + raw_datasets = load_dataset( + extension, + data_files=data_files, + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + **dataset_args, + ) + # If no validation data is there, validation_split_percentage will be used to divide the dataset. + if "validation" not in raw_datasets.keys(): + raw_datasets["validation"] = load_dataset( + extension, + data_files=data_files, + split=f"train[:{data_args.validation_split_percentage}%]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + **dataset_args, + ) + raw_datasets["train"] = load_dataset( + extension, + data_files=data_files, + split=f"train[{data_args.validation_split_percentage}%:]", + cache_dir=model_args.cache_dir, + use_auth_token=True if model_args.use_auth_token else None, + **dataset_args, + ) + + # See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at + # https://huggingface.co/docs/datasets/loading_datasets.html. + + # Load pretrained model and tokenizer + # + # Distributed training: + # The .from_pretrained methods guarantee that only one local process can concurrently + # download model & vocab. + + config_kwargs = { + "cache_dir": model_args.cache_dir, + "revision": model_args.model_revision, + "use_auth_token": True if model_args.use_auth_token else None, + } + if model_args.config_name: + config = AutoConfig.from_pretrained(model_args.config_name, **config_kwargs) + elif model_args.model_name_or_path: + config = AutoConfig.from_pretrained( + model_args.model_name_or_path, **config_kwargs + ) + else: + config = CONFIG_MAPPING[model_args.model_type]() + logger.warning("You are instantiating a new config instance from scratch.") + if model_args.config_overrides is not None: + logger.info(f"Overriding config: {model_args.config_overrides}") + config.update_from_string(model_args.config_overrides) + logger.info(f"New config: {config}") + + tokenizer_kwargs = { + "cache_dir": model_args.cache_dir, + "use_fast": model_args.use_fast_tokenizer, + "revision": model_args.model_revision, + "use_auth_token": True if model_args.use_auth_token else None, + } + if model_args.tokenizer_name: + tokenizer = AutoTokenizer.from_pretrained( + model_args.tokenizer_name, **tokenizer_kwargs + ) + elif model_args.model_name_or_path: + tokenizer = AutoTokenizer.from_pretrained( + model_args.model_name_or_path, **tokenizer_kwargs + ) + else: + raise ValueError( + "You are instantiating a new tokenizer from scratch. This is not supported by this script." + "You can do it from another script, save it, and load it from here, using --tokenizer_name." + ) + + if model_args.model_name_or_path: + torch_dtype = ( + model_args.torch_dtype + if model_args.torch_dtype in ["auto", None] + else getattr(torch, model_args.torch_dtype) + ) + model = AutoModelForCausalLM.from_pretrained( + model_args.model_name_or_path, + from_tf=bool(".ckpt" in model_args.model_name_or_path), + config=config, + cache_dir=model_args.cache_dir, + revision=model_args.model_revision, + use_auth_token=True if model_args.use_auth_token else None, + torch_dtype=torch_dtype, + low_cpu_mem_usage=model_args.low_cpu_mem_usage, + ) + else: + model = AutoModelForCausalLM.from_config(config) + n_params = sum({p.data_ptr(): p.numel() for p in model.parameters()}.values()) + logger.info( + f"Training new model from scratch - Total size={n_params/2**20:.2f}M params" + ) + + # We resize the embeddings only when necessary to avoid index errors. If you are creating a model from scratch + # on a small vocab and want a smaller embedding size, remove this test. + embedding_size = model.get_input_embeddings().weight.shape[0] + if len(tokenizer) > embedding_size: + model.resize_token_embeddings(len(tokenizer)) + + # Preprocessing the datasets. + # First we tokenize all the texts. + if training_args.do_train: + column_names = list(raw_datasets["train"].features) + else: + column_names = list(raw_datasets["validation"].features) + text_column_name = "text" if "text" in column_names else column_names[0] + + # since this will be pickled to avoid _LazyModule error in Hasher force logger loading before tokenize_function + tok_logger = transformers.utils.logging.get_logger( + "transformers.tokenization_utils_base" + ) + + def tokenize_function(examples): + with CaptureLogger(tok_logger) as cl: + output = tokenizer(examples[text_column_name]) + # clm input could be much much longer than block_size + if "Token indices sequence length is longer than the" in cl.out: + tok_logger.warning( + "^^^^^^^^^^^^^^^^ Please ignore the warning above - this long input will be chunked into smaller bits" + " before being passed to the model." + ) + return output + + with training_args.main_process_first(desc="dataset map tokenization"): + if not data_args.streaming: + tokenized_datasets = raw_datasets.map( + tokenize_function, + batched=True, + num_proc=data_args.preprocessing_num_workers, + remove_columns=column_names, + load_from_cache_file=not data_args.overwrite_cache, + desc="Running tokenizer on dataset", + ) + else: + tokenized_datasets = raw_datasets.map( + tokenize_function, + batched=True, + remove_columns=column_names, + ) + + if data_args.block_size is None: + block_size = tokenizer.model_max_length + if block_size > 1024: + logger.warning( + "The chosen tokenizer supports a `model_max_length` that is longer than the default `block_size` value" + " of 1024. If you would like to use a longer `block_size` up to `tokenizer.model_max_length` you can" + " override this default with `--block_size xxx`." + ) + block_size = 1024 + else: + if data_args.block_size > tokenizer.model_max_length: + logger.warning( + f"The block_size passed ({data_args.block_size}) is larger than the maximum length for the model" + f"({tokenizer.model_max_length}). Using block_size={tokenizer.model_max_length}." + ) + block_size = min(data_args.block_size, tokenizer.model_max_length) + + # Main data processing function that will concatenate all texts from our dataset and generate chunks of block_size. + def group_texts(examples): + # Concatenate all texts. + concatenated_examples = {k: list(chain(*examples[k])) for k in examples.keys()} + total_length = len(concatenated_examples[list(examples.keys())[0]]) + # We drop the small remainder, we could add padding if the model supported it instead of this drop, you can + # customize this part to your needs. + if total_length >= block_size: + total_length = (total_length // block_size) * block_size + # Split by chunks of max_len. + result = { + k: [t[i : i + block_size] for i in range(0, total_length, block_size)] + for k, t in concatenated_examples.items() + } + result["labels"] = result["input_ids"].copy() + return result + + # Note that with `batched=True`, this map processes 1,000 texts together, so group_texts throws away a remainder + # for each of those groups of 1,000 texts. You can adjust that batch_size here but a higher value might be slower + # to preprocess. + # + # To speed up this part, we use multiprocessing. See the documentation of the map method for more information: + # https://huggingface.co/docs/datasets/package_reference/main_classes.html#datasets.Dataset.map + + with training_args.main_process_first(desc="grouping texts together"): + if not data_args.streaming: + lm_datasets = tokenized_datasets.map( + group_texts, + batched=True, + num_proc=data_args.preprocessing_num_workers, + load_from_cache_file=not data_args.overwrite_cache, + desc=f"Grouping texts in chunks of {block_size}", + ) + else: + lm_datasets = tokenized_datasets.map( + group_texts, + batched=True, + ) + + if training_args.do_train: + if "train" not in tokenized_datasets: + raise ValueError("--do_train requires a train dataset") + train_dataset = lm_datasets["train"] + if data_args.max_train_samples is not None: + max_train_samples = min(len(train_dataset), data_args.max_train_samples) + train_dataset = train_dataset.select(range(max_train_samples)) + + if training_args.do_eval: + if "validation" not in tokenized_datasets: + raise ValueError("--do_eval requires a validation dataset") + eval_dataset = lm_datasets["validation"] + if data_args.max_eval_samples is not None: + max_eval_samples = min(len(eval_dataset), data_args.max_eval_samples) + eval_dataset = eval_dataset.select(range(max_eval_samples)) + + def preprocess_logits_for_metrics(logits, labels): + if isinstance(logits, tuple): + # Depending on the model and config, logits may contain extra tensors, + # like past_key_values, but logits always come first + logits = logits[0] + return logits.argmax(dim=-1) + + metric = evaluate.load("accuracy") + + def compute_metrics(eval_preds): + preds, labels = eval_preds + # preds have the same shape as the labels, after the argmax(-1) has been calculated + # by preprocess_logits_for_metrics but we need to shift the labels + labels = labels[:, 1:].reshape(-1) + preds = preds[:, :-1].reshape(-1) + return metric.compute(predictions=preds, references=labels) + + # Initialize our Trainer + trainer = Trainer( + model=model, + args=training_args, + train_dataset=train_dataset if training_args.do_train else None, + eval_dataset=eval_dataset if training_args.do_eval else None, + tokenizer=tokenizer, + # Data collator will default to DataCollatorWithPadding, so we change it. + data_collator=default_data_collator, + compute_metrics=compute_metrics + if training_args.do_eval and not is_torch_tpu_available() + else None, + preprocess_logits_for_metrics=preprocess_logits_for_metrics + if training_args.do_eval and not is_torch_tpu_available() + else None, + ) + + # Training + if training_args.do_train: + checkpoint = None + if training_args.resume_from_checkpoint is not None: + checkpoint = training_args.resume_from_checkpoint + elif last_checkpoint is not None: + checkpoint = last_checkpoint + train_result = trainer.train(resume_from_checkpoint=checkpoint) + trainer.save_model() # Saves the tokenizer too for easy upload + + metrics = train_result.metrics + + max_train_samples = ( + data_args.max_train_samples + if data_args.max_train_samples is not None + else len(train_dataset) + ) + metrics["train_samples"] = min(max_train_samples, len(train_dataset)) + + trainer.log_metrics("train", metrics) + trainer.save_metrics("train", metrics) + trainer.save_state() + + # Evaluation + if training_args.do_eval: + logger.info("*** Evaluate ***") + + metrics = trainer.evaluate() + + max_eval_samples = ( + data_args.max_eval_samples + if data_args.max_eval_samples is not None + else len(eval_dataset) + ) + metrics["eval_samples"] = min(max_eval_samples, len(eval_dataset)) + try: + perplexity = math.exp(metrics["eval_loss"]) + except OverflowError: + perplexity = float("inf") + metrics["perplexity"] = perplexity + + trainer.log_metrics("eval", metrics) + trainer.save_metrics("eval", metrics) + + kwargs = { + "finetuned_from": model_args.model_name_or_path, + "tasks": "text-generation", + } + if data_args.dataset_name is not None: + kwargs["dataset_tags"] = data_args.dataset_name + if data_args.dataset_config_name is not None: + kwargs["dataset_args"] = data_args.dataset_config_name + kwargs[ + "dataset" + ] = f"{data_args.dataset_name} {data_args.dataset_config_name}" + else: + kwargs["dataset"] = data_args.dataset_name + + if training_args.push_to_hub: + trainer.push_to_hub(**kwargs) + else: + trainer.create_model_card(**kwargs) + + +def _mp_fn(index): + # For xla_spawn (TPUs) + main() + + +if __name__ == "__main__": + main() diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/gptfttest.yaml b/demo-notebooks/guided-demos/notebook-ex-outputs/gptfttest.yaml new file mode 100644 index 000000000..54799e10c --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/gptfttest.yaml @@ -0,0 +1,175 @@ +apiVersion: mcad.ibm.com/v1beta1 +kind: AppWrapper +metadata: + labels: + orderedinstance: m5.xlarge_g4dn.xlarge + name: gptfttest + namespace: default +spec: + priority: 9 + resources: + GenericItems: + - custompodresources: + - limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + replicas: 1 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + - limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + replicas: 2 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + generictemplate: + apiVersion: ray.io/v1alpha1 + kind: RayCluster + metadata: + labels: + appwrapper.mcad.ibm.com: gptfttest + controller-tools.k8s.io: '1.0' + name: gptfttest + namespace: default + spec: + autoscalerOptions: + idleTimeoutSeconds: 60 + imagePullPolicy: Always + resources: + limits: + cpu: 500m + memory: 512Mi + requests: + cpu: 500m + memory: 512Mi + upscalingMode: Default + enableInTreeAutoscaling: false + headGroupSpec: + rayStartParams: + block: 'true' + dashboard-host: 0.0.0.0 + num-gpus: '0' + serviceType: ClusterIP + template: + spec: + affinity: + nodeAffinity: + requiredDuringSchedulingIgnoredDuringExecution: + nodeSelectorTerms: + - matchExpressions: + - key: gptfttest + operator: In + values: + - gptfttest + containers: + - env: + - name: MY_POD_IP + valueFrom: + fieldRef: + fieldPath: status.podIP + image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + imagePullPolicy: Always + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: ray-head + ports: + - containerPort: 6379 + name: gcs + - containerPort: 8265 + name: dashboard + - containerPort: 10001 + name: client + resources: + limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + rayVersion: 1.12.0 + workerGroupSpecs: + - groupName: small-group-gptfttest + maxReplicas: 2 + minReplicas: 2 + rayStartParams: + block: 'true' + num-gpus: '1' + replicas: 2 + template: + metadata: + annotations: + key: value + labels: + key: value + spec: + affinity: + nodeAffinity: + requiredDuringSchedulingIgnoredDuringExecution: + nodeSelectorTerms: + - matchExpressions: + - key: gptfttest + operator: In + values: + - gptfttest + containers: + - env: + - name: MY_POD_IP + valueFrom: + fieldRef: + fieldPath: status.podIP + image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: machine-learning + resources: + limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + initContainers: + - command: + - sh + - -c + - until nslookup $RAY_IP.$(cat /var/run/secrets/kubernetes.io/serviceaccount/namespace).svc.cluster.local; + do echo waiting for myservice; sleep 2; done + image: busybox:1.28 + name: init-myservice + replicas: 1 + - generictemplate: + apiVersion: route.openshift.io/v1 + kind: Route + metadata: + labels: + odh-ray-cluster-service: gptfttest-head-svc + name: ray-dashboard-gptfttest + namespace: default + spec: + port: + targetPort: dashboard + to: + kind: Service + name: gptfttest-head-svc + replica: 1 + Items: [] diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/instascaletest.yaml b/demo-notebooks/guided-demos/notebook-ex-outputs/instascaletest.yaml new file mode 100644 index 000000000..2a40438cf --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/instascaletest.yaml @@ -0,0 +1,170 @@ +apiVersion: mcad.ibm.com/v1beta1 +kind: AppWrapper +metadata: + labels: + orderedinstance: m5.xlarge_g4dn.xlarge + name: instascaletest + namespace: default +spec: + priority: 9 + resources: + GenericItems: + - custompodresources: + - limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + replicas: 1 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + - limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + replicas: 2 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + generictemplate: + apiVersion: ray.io/v1alpha1 + kind: RayCluster + metadata: + labels: + appwrapper.mcad.ibm.com: instascaletest + controller-tools.k8s.io: '1.0' + name: instascaletest + namespace: default + spec: + autoscalerOptions: + idleTimeoutSeconds: 60 + imagePullPolicy: Always + resources: + limits: + cpu: 500m + memory: 512Mi + requests: + cpu: 500m + memory: 512Mi + upscalingMode: Default + enableInTreeAutoscaling: false + headGroupSpec: + rayStartParams: + block: 'true' + dashboard-host: 0.0.0.0 + num-gpus: '0' + serviceType: ClusterIP + template: + spec: + affinity: + nodeAffinity: + requiredDuringSchedulingIgnoredDuringExecution: + nodeSelectorTerms: + - matchExpressions: + - key: instascaletest + operator: In + values: + - instascaletest + containers: + - image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + imagePullPolicy: Always + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: ray-head + ports: + - containerPort: 6379 + name: gcs + - containerPort: 8265 + name: dashboard + - containerPort: 10001 + name: client + resources: + limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + rayVersion: 1.12.0 + workerGroupSpecs: + - groupName: small-group-instascaletest + maxReplicas: 2 + minReplicas: 2 + rayStartParams: + block: 'true' + num-gpus: '1' + replicas: 2 + template: + metadata: + annotations: + key: value + labels: + key: value + spec: + affinity: + nodeAffinity: + requiredDuringSchedulingIgnoredDuringExecution: + nodeSelectorTerms: + - matchExpressions: + - key: instascaletest + operator: In + values: + - instascaletest + containers: + - env: + - name: MY_POD_IP + valueFrom: + fieldRef: + fieldPath: status.podIP + image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: machine-learning + resources: + limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + initContainers: + - command: + - sh + - -c + - until nslookup $RAY_IP.$(cat /var/run/secrets/kubernetes.io/serviceaccount/namespace).svc.cluster.local; + do echo waiting for myservice; sleep 2; done + image: busybox:1.28 + name: init-myservice + replicas: 1 + - generictemplate: + apiVersion: route.openshift.io/v1 + kind: Route + metadata: + labels: + odh-ray-cluster-service: instascaletest-head-svc + name: ray-dashboard-instascaletest + namespace: default + spec: + port: + targetPort: dashboard + to: + kind: Service + name: instascaletest-head-svc + replica: 1 + Items: [] diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/interactivetest.yaml b/demo-notebooks/guided-demos/notebook-ex-outputs/interactivetest.yaml new file mode 100644 index 000000000..5544dfe72 --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/interactivetest.yaml @@ -0,0 +1,175 @@ +apiVersion: mcad.ibm.com/v1beta1 +kind: AppWrapper +metadata: + labels: + orderedinstance: m5.xlarge_g4dn.xlarge + name: interactivetest + namespace: default +spec: + priority: 9 + resources: + GenericItems: + - custompodresources: + - limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + replicas: 1 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + - limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + replicas: 2 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + generictemplate: + apiVersion: ray.io/v1alpha1 + kind: RayCluster + metadata: + labels: + appwrapper.mcad.ibm.com: interactivetest + controller-tools.k8s.io: '1.0' + name: interactivetest + namespace: default + spec: + autoscalerOptions: + idleTimeoutSeconds: 60 + imagePullPolicy: Always + resources: + limits: + cpu: 500m + memory: 512Mi + requests: + cpu: 500m + memory: 512Mi + upscalingMode: Default + enableInTreeAutoscaling: false + headGroupSpec: + rayStartParams: + block: 'true' + dashboard-host: 0.0.0.0 + num-gpus: '0' + serviceType: ClusterIP + template: + spec: + affinity: + nodeAffinity: + requiredDuringSchedulingIgnoredDuringExecution: + nodeSelectorTerms: + - matchExpressions: + - key: interactivetest + operator: In + values: + - interactivetest + containers: + - env: + - name: MY_POD_IP + valueFrom: + fieldRef: + fieldPath: status.podIP + image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + imagePullPolicy: Always + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: ray-head + ports: + - containerPort: 6379 + name: gcs + - containerPort: 8265 + name: dashboard + - containerPort: 10001 + name: client + resources: + limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + rayVersion: 1.12.0 + workerGroupSpecs: + - groupName: small-group-interactivetest + maxReplicas: 2 + minReplicas: 2 + rayStartParams: + block: 'true' + num-gpus: '1' + replicas: 2 + template: + metadata: + annotations: + key: value + labels: + key: value + spec: + affinity: + nodeAffinity: + requiredDuringSchedulingIgnoredDuringExecution: + nodeSelectorTerms: + - matchExpressions: + - key: interactivetest + operator: In + values: + - interactivetest + containers: + - env: + - name: MY_POD_IP + valueFrom: + fieldRef: + fieldPath: status.podIP + image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: machine-learning + resources: + limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 1 + initContainers: + - command: + - sh + - -c + - until nslookup $RAY_IP.$(cat /var/run/secrets/kubernetes.io/serviceaccount/namespace).svc.cluster.local; + do echo waiting for myservice; sleep 2; done + image: busybox:1.28 + name: init-myservice + replicas: 1 + - generictemplate: + apiVersion: route.openshift.io/v1 + kind: Route + metadata: + labels: + odh-ray-cluster-service: interactivetest-head-svc + name: ray-dashboard-interactivetest + namespace: default + spec: + port: + targetPort: dashboard + to: + kind: Service + name: interactivetest-head-svc + replica: 1 + Items: [] diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/jobtest.yaml b/demo-notebooks/guided-demos/notebook-ex-outputs/jobtest.yaml new file mode 100644 index 000000000..d08517256 --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/jobtest.yaml @@ -0,0 +1,155 @@ +apiVersion: mcad.ibm.com/v1beta1 +kind: AppWrapper +metadata: + name: jobtest + namespace: default +spec: + priority: 9 + resources: + GenericItems: + - custompodresources: + - limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + replicas: 1 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + - limits: + cpu: 1 + memory: 4G + nvidia.com/gpu: 0 + replicas: 2 + requests: + cpu: 1 + memory: 4G + nvidia.com/gpu: 0 + generictemplate: + apiVersion: ray.io/v1alpha1 + kind: RayCluster + metadata: + labels: + appwrapper.mcad.ibm.com: jobtest + controller-tools.k8s.io: '1.0' + name: jobtest + namespace: default + spec: + autoscalerOptions: + idleTimeoutSeconds: 60 + imagePullPolicy: Always + resources: + limits: + cpu: 500m + memory: 512Mi + requests: + cpu: 500m + memory: 512Mi + upscalingMode: Default + enableInTreeAutoscaling: false + headGroupSpec: + rayStartParams: + block: 'true' + dashboard-host: 0.0.0.0 + num-gpus: '0' + serviceType: ClusterIP + template: + spec: + containers: + - env: + - name: MY_POD_IP + valueFrom: + fieldRef: + fieldPath: status.podIP + image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + imagePullPolicy: Always + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: ray-head + ports: + - containerPort: 6379 + name: gcs + - containerPort: 8265 + name: dashboard + - containerPort: 10001 + name: client + resources: + limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + rayVersion: 1.12.0 + workerGroupSpecs: + - groupName: small-group-jobtest + maxReplicas: 2 + minReplicas: 2 + rayStartParams: + block: 'true' + num-gpus: '0' + replicas: 2 + template: + metadata: + annotations: + key: value + labels: + key: value + spec: + containers: + - env: + - name: MY_POD_IP + valueFrom: + fieldRef: + fieldPath: status.podIP + image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: machine-learning + resources: + limits: + cpu: 1 + memory: 4G + nvidia.com/gpu: 0 + requests: + cpu: 1 + memory: 4G + nvidia.com/gpu: 0 + initContainers: + - command: + - sh + - -c + - until nslookup $RAY_IP.$(cat /var/run/secrets/kubernetes.io/serviceaccount/namespace).svc.cluster.local; + do echo waiting for myservice; sleep 2; done + image: busybox:1.28 + name: init-myservice + replicas: 1 + - generictemplate: + apiVersion: route.openshift.io/v1 + kind: Route + metadata: + labels: + odh-ray-cluster-service: jobtest-head-svc + name: ray-dashboard-jobtest + namespace: default + spec: + port: + targetPort: dashboard + to: + kind: Service + name: jobtest-head-svc + replica: 1 + Items: [] diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/mnist.py b/demo-notebooks/guided-demos/notebook-ex-outputs/mnist.py new file mode 100644 index 000000000..6eb663dc7 --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/mnist.py @@ -0,0 +1,160 @@ +# Copyright 2022 IBM, Red Hat +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +# In[] +import os + +import torch +from pytorch_lightning import LightningModule, Trainer +from pytorch_lightning.callbacks.progress import TQDMProgressBar +from pytorch_lightning.loggers import CSVLogger +from torch import nn +from torch.nn import functional as F +from torch.utils.data import DataLoader, random_split +from torchmetrics import Accuracy +from torchvision import transforms +from torchvision.datasets import MNIST + +PATH_DATASETS = os.environ.get("PATH_DATASETS", ".") +BATCH_SIZE = 256 if torch.cuda.is_available() else 64 +# %% + +print("prior to running the trainer") +print("MASTER_ADDR: is ", os.getenv("MASTER_ADDR")) +print("MASTER_PORT: is ", os.getenv("MASTER_PORT")) + + +class LitMNIST(LightningModule): + def __init__(self, data_dir=PATH_DATASETS, hidden_size=64, learning_rate=2e-4): + super().__init__() + + # Set our init args as class attributes + self.data_dir = data_dir + self.hidden_size = hidden_size + self.learning_rate = learning_rate + + # Hardcode some dataset specific attributes + self.num_classes = 10 + self.dims = (1, 28, 28) + channels, width, height = self.dims + self.transform = transforms.Compose( + [ + transforms.ToTensor(), + transforms.Normalize((0.1307,), (0.3081,)), + ] + ) + + # Define PyTorch model + self.model = nn.Sequential( + nn.Flatten(), + nn.Linear(channels * width * height, hidden_size), + nn.ReLU(), + nn.Dropout(0.1), + nn.Linear(hidden_size, hidden_size), + nn.ReLU(), + nn.Dropout(0.1), + nn.Linear(hidden_size, self.num_classes), + ) + + self.val_accuracy = Accuracy() + self.test_accuracy = Accuracy() + + def forward(self, x): + x = self.model(x) + return F.log_softmax(x, dim=1) + + def training_step(self, batch, batch_idx): + x, y = batch + logits = self(x) + loss = F.nll_loss(logits, y) + return loss + + def validation_step(self, batch, batch_idx): + x, y = batch + logits = self(x) + loss = F.nll_loss(logits, y) + preds = torch.argmax(logits, dim=1) + self.val_accuracy.update(preds, y) + + # Calling self.log will surface up scalars for you in TensorBoard + self.log("val_loss", loss, prog_bar=True) + self.log("val_acc", self.val_accuracy, prog_bar=True) + + def test_step(self, batch, batch_idx): + x, y = batch + logits = self(x) + loss = F.nll_loss(logits, y) + preds = torch.argmax(logits, dim=1) + self.test_accuracy.update(preds, y) + + # Calling self.log will surface up scalars for you in TensorBoard + self.log("test_loss", loss, prog_bar=True) + self.log("test_acc", self.test_accuracy, prog_bar=True) + + def configure_optimizers(self): + optimizer = torch.optim.Adam(self.parameters(), lr=self.learning_rate) + return optimizer + + #################### + # DATA RELATED HOOKS + #################### + + def prepare_data(self): + # download + print("Downloading MNIST dataset...") + MNIST(self.data_dir, train=True, download=True) + MNIST(self.data_dir, train=False, download=True) + + def setup(self, stage=None): + # Assign train/val datasets for use in dataloaders + if stage == "fit" or stage is None: + mnist_full = MNIST(self.data_dir, train=True, transform=self.transform) + self.mnist_train, self.mnist_val = random_split(mnist_full, [55000, 5000]) + + # Assign test dataset for use in dataloader(s) + if stage == "test" or stage is None: + self.mnist_test = MNIST( + self.data_dir, train=False, transform=self.transform + ) + + def train_dataloader(self): + return DataLoader(self.mnist_train, batch_size=BATCH_SIZE) + + def val_dataloader(self): + return DataLoader(self.mnist_val, batch_size=BATCH_SIZE) + + def test_dataloader(self): + return DataLoader(self.mnist_test, batch_size=BATCH_SIZE) + + +# Init DataLoader from MNIST Dataset + +model = LitMNIST() + +print("GROUP: ", int(os.environ.get("GROUP_WORLD_SIZE", 1))) +print("LOCAL: ", int(os.environ.get("LOCAL_WORLD_SIZE", 1))) + +# Initialize a trainer +trainer = Trainer( + accelerator="auto", + # devices=1 if torch.cuda.is_available() else None, # limiting got iPython runs + max_epochs=5, + callbacks=[TQDMProgressBar(refresh_rate=20)], + num_nodes=int(os.environ.get("GROUP_WORLD_SIZE", 1)), + devices=int(os.environ.get("LOCAL_WORLD_SIZE", 1)), + strategy="ddp", +) + +# Train the model ⚡ +trainer.fit(model) diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/raytest.yaml b/demo-notebooks/guided-demos/notebook-ex-outputs/raytest.yaml new file mode 100644 index 000000000..c3e804dd4 --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/raytest.yaml @@ -0,0 +1,155 @@ +apiVersion: mcad.ibm.com/v1beta1 +kind: AppWrapper +metadata: + name: raytest + namespace: default +spec: + priority: 9 + resources: + GenericItems: + - custompodresources: + - limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + replicas: 1 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + - limits: + cpu: 1 + memory: 4G + nvidia.com/gpu: 0 + replicas: 2 + requests: + cpu: 1 + memory: 4G + nvidia.com/gpu: 0 + generictemplate: + apiVersion: ray.io/v1alpha1 + kind: RayCluster + metadata: + labels: + appwrapper.mcad.ibm.com: raytest + controller-tools.k8s.io: '1.0' + name: raytest + namespace: default + spec: + autoscalerOptions: + idleTimeoutSeconds: 60 + imagePullPolicy: Always + resources: + limits: + cpu: 500m + memory: 512Mi + requests: + cpu: 500m + memory: 512Mi + upscalingMode: Default + enableInTreeAutoscaling: false + headGroupSpec: + rayStartParams: + block: 'true' + dashboard-host: 0.0.0.0 + num-gpus: '0' + serviceType: ClusterIP + template: + spec: + containers: + - env: + - name: MY_POD_IP + valueFrom: + fieldRef: + fieldPath: status.podIP + image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + imagePullPolicy: Always + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: ray-head + ports: + - containerPort: 6379 + name: gcs + - containerPort: 8265 + name: dashboard + - containerPort: 10001 + name: client + resources: + limits: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + requests: + cpu: 2 + memory: 8G + nvidia.com/gpu: 0 + rayVersion: 1.12.0 + workerGroupSpecs: + - groupName: small-group-raytest + maxReplicas: 2 + minReplicas: 2 + rayStartParams: + block: 'true' + num-gpus: '0' + replicas: 2 + template: + metadata: + annotations: + key: value + labels: + key: value + spec: + containers: + - env: + - name: MY_POD_IP + valueFrom: + fieldRef: + fieldPath: status.podIP + image: ghcr.io/foundation-model-stack/base:ray2.1.0-py38-gpu-pytorch1.12.0cu116-20221213-193103 + lifecycle: + preStop: + exec: + command: + - /bin/sh + - -c + - ray stop + name: machine-learning + resources: + limits: + cpu: 1 + memory: 4G + nvidia.com/gpu: 0 + requests: + cpu: 1 + memory: 4G + nvidia.com/gpu: 0 + initContainers: + - command: + - sh + - -c + - until nslookup $RAY_IP.$(cat /var/run/secrets/kubernetes.io/serviceaccount/namespace).svc.cluster.local; + do echo waiting for myservice; sleep 2; done + image: busybox:1.28 + name: init-myservice + replicas: 1 + - generictemplate: + apiVersion: route.openshift.io/v1 + kind: Route + metadata: + labels: + odh-ray-cluster-service: raytest-head-svc + name: ray-dashboard-raytest + namespace: default + spec: + port: + targetPort: dashboard + to: + kind: Service + name: raytest-head-svc + replica: 1 + Items: [] diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/requirements.txt b/demo-notebooks/guided-demos/notebook-ex-outputs/requirements.txt new file mode 100644 index 000000000..7266b064a --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/requirements.txt @@ -0,0 +1,4 @@ +pytorch_lightning==1.5.10 +ray_lightning +torchmetrics==0.9.1 +torchvision==0.12.0 diff --git a/demo-notebooks/guided-demos/notebook-ex-outputs/requirements_gpt.txt b/demo-notebooks/guided-demos/notebook-ex-outputs/requirements_gpt.txt new file mode 100644 index 000000000..bd6c4f525 --- /dev/null +++ b/demo-notebooks/guided-demos/notebook-ex-outputs/requirements_gpt.txt @@ -0,0 +1,8 @@ +accelerate >= 0.12.0 +torch >= 1.3 +datasets >= 1.8.0 +sentencepiece != 0.1.92 +evaluate +scikit-learn +transformers==4.28.1 +protobuf<=3.20.1,>=3.8.0 diff --git a/demo-notebooks/guided-demos/requirements.txt b/demo-notebooks/guided-demos/requirements.txt new file mode 100644 index 000000000..7266b064a --- /dev/null +++ b/demo-notebooks/guided-demos/requirements.txt @@ -0,0 +1,4 @@ +pytorch_lightning==1.5.10 +ray_lightning +torchmetrics==0.9.1 +torchvision==0.12.0 diff --git a/demo-notebooks/guided-demos/requirements_gpt.txt b/demo-notebooks/guided-demos/requirements_gpt.txt new file mode 100644 index 000000000..bd6c4f525 --- /dev/null +++ b/demo-notebooks/guided-demos/requirements_gpt.txt @@ -0,0 +1,8 @@ +accelerate >= 0.12.0 +torch >= 1.3 +datasets >= 1.8.0 +sentencepiece != 0.1.92 +evaluate +scikit-learn +transformers==4.28.1 +protobuf<=3.20.1,>=3.8.0