Sync with upstream master to pull latest changes

Merge remote-tracking branch 'upstream/master' into feat/webhook-validate-trainjob-name

Author: juniemariam

.github/workflows/gh-workflow-approve.yaml

@@ -0,0 +1,72 @@
+name: Approve Workflow Runs
+
+permissions:
+  actions: write
+  contents: read
+
+on:
+  pull_request_target:
+    types:
+      - labeled
+      - synchronize
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.ref }}-${{ github.event.number }}
+  cancel-in-progress: true
+
+jobs:
+  ok-to-test:
+    if: contains(github.event.pull_request.labels.*.name, 'ok-to-test') || github.event_name == 'pull_request_target'
+    runs-on: ubuntu-latest
+    continue-on-error: true
+
+    steps:
+      - name: Check if author is a Kubeflow GitHub member
+        id: membership-check
+        uses: actions/github-script@v7
+        with:
+          script: |
+            const username = context.payload.pull_request.user.login;
+            const org = context.repo.owner;
+            try {
+              const res = await github.rest.orgs.checkMembershipForUser({
+                org,
+                username
+              });
+              core.setOutput("is_member", true);
+            } catch (error) {
+              if (error.status === 404) {
+                // User is not a member
+                core.setOutput("is_member", false);
+              } else {
+                throw error;
+              }
+            }
+
+      - name: Approve Pending Workflow Runs
+        if: steps.membership-check.outputs.is_member == 'true' || contains(github.event.pull_request.labels.*.name, 'ok-to-test')
+        uses: actions/github-script@v7
+        with:
+          retries: 3
+          script: |
+            const request = {
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              event: "pull_request",
+              status: "action_required",
+              head_sha: context.payload.pull_request.head.sha,
+            }
+
+            core.info(`Getting workflow runs that need approval for commit ${request.head_sha}`)
+            const runs = await github.paginate(github.rest.actions.listWorkflowRunsForRepo, request)
+
+            core.info(`Found ${runs.length} workflow runs that need approval`)
+            for (const run of runs) {
+              core.info(`Approving workflow run ${run.id}`)
+              const request = {
+                owner: context.repo.owner,
+                repo: context.repo.repo,
+                run_id: run.id,
+              }
+              await github.rest.actions.approveWorkflowRun(request)
+            }

.github/workflows/test-e2e.yaml

@@ -41,7 +41,7 @@ jobs:
           pip install papermill==2.6.0 jupyter==1.1.1 ipykernel==6.29.5
 
           echo "Install Kubeflow SDK"
-          pip install git+https://github.com/kubeflow/sdk.git@main#subdirectory=python
+          pip install git+https://github.com/kubeflow/sdk.git@main
 
       - name: Setup cluster
         run: |

.github/workflows/test-go.yaml

@@ -77,12 +77,3 @@ jobs:
         with:
           path-to-profile: cover.out
           working-directory: ${{ env.GOPATH }}/src/github.com/kubeflow/trainer
-          parallel: true
-
-  finish:
-    needs: test
-    runs-on: ubuntu-latest
-    steps:
-      - uses: shogo82148/actions-goveralls@v1
-        with:
-          parallel-finished: true

.gitignore

@@ -16,9 +16,10 @@ __debug_bin
 # The default output for various artifacts (e.g. Jupyter Notebooks after Papermill execution).
 artifacts
 
-# Python cache files
+# Python cache files / packaging
 __pycache__/
 *.egg-info/
+dist/
 
 # Coverage
 cover.out

CONTRIBUTING.md

@@ -135,4 +135,4 @@ On ubuntu the default go package appears to be gccgo-go which has problems. It's
 
 Changes to the Kubeflow Trainer Python SDK can be made in the https://github.com/kubeflow/sdk repo.
 
-The Trainer SDK can be found at https://github.com/kubeflow/sdk/tree/main/python/kubeflow/trainer.
+The Trainer SDK can be found at https://github.com/kubeflow/sdk/tree/main/kubeflow/trainer.

Makefile

@@ -152,11 +152,11 @@ endif
 # Instructions to run tests.
 .PHONY: test
 test: ## Run Go unit test.
-	go test $(shell go list ./... | grep -v '/test/' | grep -v '/cmd/' | grep -v '/hack/' | grep -v '/pkg/apis' | grep -v '/pkg/client') -coverprofile cover.out
+	go test $(shell go list ./... | grep -Ev '/(test|cmd|hack|pkg/apis|pkg/client|pkg/util/testing)') -coverprofile cover.out
 
 .PHONY: test-integration
 test-integration: ginkgo envtest jobset-operator-crd scheduler-plugins-crd ## Run Go integration test.
-	KUBEBUILDER_ASSETS="$(shell $(ENVTEST) use $(K8S_VERSION) -p path)" $(GINKGO) -coverprofile cover.out -v ./test/integration/...
+	KUBEBUILDER_ASSETS="$(shell $(ENVTEST) use $(K8S_VERSION) -p path)" $(GINKGO) -v ./test/integration/...
 
 .PHONY: test-python
 test-python: ## Run Python unit test.

README.md

@@ -1,15 +1,26 @@
 # Kubeflow Trainer
 
-[![Build Status](https://github.com/kubeflow/trainer/actions/workflows/test-go.yaml/badge.svg?branch=master)](https://github.com/kubeflow/trainer/actions/workflows/test-go.yaml?branch=master)
+[![Join Slack](https://img.shields.io/badge/Join_Slack-blue?logo=slack)](https://www.kubeflow.org/docs/about/community/#kubeflow-slack-channels)
 [![Coverage Status](https://coveralls.io/repos/github/kubeflow/trainer/badge.svg?branch=master)](https://coveralls.io/github/kubeflow/trainer?branch=master)
 [![Go Report Card](https://goreportcard.com/badge/github.com/kubeflow/trainer)](https://goreportcard.com/report/github.com/kubeflow/trainer)
 [![OpenSSF Best Practices](https://www.bestpractices.dev/projects/10435/badge)](https://www.bestpractices.dev/projects/10435)
+[![Ask DeepWiki](https://deepwiki.com/badge.svg)](https://deepwiki.com/kubeflow/trainer)
 
 <h1 align="center">
     <img src="./docs/images/trainer-logo.svg" alt="logo" width="200">
   <br>
 </h1>
 
+Latest News 🔥
+
+- [2025/07] PyTorch on Kubernetes: Kubeflow Trainer Joins the PyTorch Ecosystem. Find the
+  announcement in [the PyTorch blog post](https://pytorch.org/blog/pytorch-on-kubernetes-kubeflow-trainer-joins-the-pytorch-ecosystem/).
+- [2025/07] Kubeflow Trainer v2.0 has been officially released. Check out
+  [the blog post announcement](https://blog.kubeflow.org/trainer/intro/) and [the
+  release notes](https://github.com/kubeflow/trainer/releases/tag/v2.0.0).
+- [2025/04] From High Performance Computing To AI Workloads on Kubernetes: MPI Runtime in
+  Kubeflow TrainJob. See the [KubeCon + CloudNativeCon London talk](https://youtu.be/Fnb1a5Kaxgo)
+
 ## Overview
 
 Kubeflow Trainer is a Kubernetes-native project designed for large language models (LLMs)
@@ -18,7 +29,7 @@ various frameworks, including PyTorch, JAX, TensorFlow, and others.
 
 You can integrate other ML libraries such as [HuggingFace](https://huggingface.co),
 [DeepSpeed](https://github.com/microsoft/DeepSpeed), or [Megatron-LM](https://github.com/NVIDIA/Megatron-LM)
-with Kubeflow Training to orchestrate their ML training on Kubernetes.
+with Kubeflow Trainer to run them on Kubernetes.
 
 Kubeflow Trainer enables you to effortlessly develop your LLMs with the
 [Kubeflow Python SDK](https://github.com/kubeflow/sdk/), and build Kubernetes-native Training
@@ -37,7 +48,7 @@ The following KubeCon + CloudNativeCon 2024 talk provides an overview of Kubeflo
 
 ## Getting Started
 
-Please check [the official Kubeflow documentation](https://www.kubeflow.org/docs/components/trainer/getting-started)
+Please check [the official Kubeflow Trainer documentation](https://www.kubeflow.org/docs/components/trainer/getting-started)
 to install and get started with Kubeflow Trainer.
 
 ## Community

api/python_api/pyproject.toml

@@ -1,22 +1,18 @@
-[build-system]
-requires = ["hatchling"]
-build-backend = "hatchling.build"
-
 [project]
 name = "kubeflow_trainer_api"
 dynamic = ["version"]
+requires-python = ">=3.9"
 authors = [
   { name = "The Kubeflow Authors", email = "kubeflow-discuss@googlegroups.com" },
 ]
-license = { file = "../../LICENSE" }
+license = "Apache-2.0"
 description = "Kubeflow Trainer API models for Kubernetes resources to interact with Kubeflow APIs."
 readme = "README.md"
 keywords = ["kubeflow", "trainer", "model training", "llm", "ai", "api"]
 classifiers = [
   "Intended Audience :: Developers",
   "Intended Audience :: Education",
   "Intended Audience :: Science/Research",
-  "Programming Language :: Python :: 3.8",
   "Programming Language :: Python :: 3.9",
   "Programming Language :: Python :: 3.10",
   "Programming Language :: Python :: 3.11",
@@ -35,6 +31,10 @@ Homepage = "https://github.com/kubeflow/trainer"
 Documentation = "https://www.kubeflow.org/docs/components/trainer/"
 Source = "https://github.com/kubeflow/trainer"
 
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
 [tool.hatch.build.targets.wheel]
 packages = ["kubeflow_trainer_api"]
 

charts/kubeflow-trainer/values.yaml

@@ -92,16 +92,14 @@ manager:
     #   memory: 300Mi
 
   # -- Security context for manager containers.
-  securityContext: {}
-    # readOnlyRootFilesystem: true
-    # privileged: false
-    # allowPrivilegeEscalation: false
-    # runAsNonRoot: true
-    # capabilities:
-    #   drop:
-    #   - ALL
-    # seccompProfile:
-    #   type: RuntimeDefault
+  securityContext:
+    allowPrivilegeEscalation: false
+    runAsNonRoot: true
+    capabilities:
+      drop:
+      - ALL
+    seccompProfile:
+      type: RuntimeDefault
 
 webhook:
   # -- Specifies how unrecognized errors are handled.

cmd/runtimes/deepspeed/requirements.txt

@@ -1,9 +1,9 @@
 # Keep the same version as for Torch runtime.
 torch==2.7.1
 # DeepSpeed libraries.
-deepspeed==0.17.1
-mpi4py==4.0.3
+deepspeed==0.17.4
+mpi4py==4.1.0
 # HuggingFace libraries.
-datasets==3.4.1
-transformers==4.50.0
+datasets==4.0.0
+transformers==4.55.0
 SentencePiece==0.2.0