Fix intent detection code issue (#651)

* update lvm tgi llama requirements

Signed-off-by: letonghan <[email protected]>

* fix intent detection issue

Signed-off-by: letonghan <[email protected]>

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Signed-off-by: letonghan <[email protected]>
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: letonghan

comps/intent_detection/langchain/intent_detection.py

@@ -31,7 +31,8 @@ def llm_generate(input: LLMParamsDoc):
         timeout=600,
     )
 
-    prompt = PromptTemplate(template=IntentTemplate.generate_intent_template, input_variables=["query"])
+    prompt_template = 'Please identify the intent of the user query. You may only respond with "chitchat" or \QA" without explanations or engaging in conversation.### User Query: {query}, ### Response: '
+    prompt = PromptTemplate(template=prompt_template, input_variables=["query"])
 
     llm_chain = LLMChain(prompt=prompt, llm=llm)
 

comps/intent_detection/langchain/requirements.txt

@@ -7,3 +7,4 @@ opentelemetry-exporter-otlp
 opentelemetry-sdk
 prometheus-fastapi-instrumentator
 shortuuid
+uvicorn

comps/intent_detection/langchain/template.py

@@ -2,6 +2,7 @@ datasets
 docarray[full]
 fastapi
 huggingface_hub
+langchain-core
 opentelemetry-api
 opentelemetry-exporter-otlp
 opentelemetry-sdk

comps/lvms/tgi-llava/requirements.txt

@@ -7,9 +7,10 @@ set -xe
 WORKPATH=$(dirname "$PWD")
 LOG_PATH="$WORKPATH/tests"
 ip_address=$(hostname -I | awk '{print $1}')
+
 function build_docker_images() {
     cd $WORKPATH
-    docker build --no-cache -t opea/llm-tgi:latest -f comps/intent_detection/langchain/Dockerfile .
+    docker build --no-cache -t opea/intent-detection:comps -f comps/intent_detection/langchain/Dockerfile .
 }
 
 function start_service() {
@@ -22,8 +23,19 @@ function start_service() {
     export TGI_LLM_ENDPOINT="http://${ip_address}:${tgi_endpoint}"
     intent_port=5043
     unset http_proxy
-    docker run -d --name="test-comps-intent-server" -p ${intent_port}:9000 --ipc=host -e http_proxy=$http_proxy -e https_proxy=$https_proxy -e TGI_LLM_ENDPOINT=$TGI_LLM_ENDPOINT -e HUGGINGFACEHUB_API_TOKEN=$HUGGINGFACEHUB_API_TOKEN opea/llm-tgi:latest
-    sleep 5m
+    docker run -d --name="test-comps-intent-server" -p ${intent_port}:9000 --ipc=host -e http_proxy=$http_proxy -e https_proxy=$https_proxy -e TGI_LLM_ENDPOINT=$TGI_LLM_ENDPOINT -e HUGGINGFACEHUB_API_TOKEN=$HUGGINGFACEHUB_API_TOKEN opea/intent-detection:comps
+
+    # check whether tgi is fully ready
+    n=0
+    until [[ "$n" -ge 100 ]] || [[ $ready == true ]]; do
+        docker logs test-comps-intent-tgi-endpoint > ${LOG_PATH}/tgi.log
+        n=$((n+1))
+        if grep -q Connected ${LOG_PATH}/tgi.log; then
+            break
+        fi
+        sleep 5s
+    done
+    sleep 5s
 }
 
 function validate_microservice() {
@@ -33,11 +45,16 @@ function validate_microservice() {
         -d '{"query":"What is Deep Learning?","max_new_tokens":10,"top_k":1,"temperature":0.001,"streaming":false}' \
         -H 'Content-Type: application/json')
 
-    echo "==============="
-    echo $result
+    if [[ $result == *"QA"* ]]; then
+        echo $result
+        echo "Result correct."
+    else
+        echo "Result wrong. Received was $result"
+        docker logs test-comps-intent-server > ${LOG_PATH}/intent_detection.log
+        docker logs test-comps-intent-tgi-endpoint > ${LOG_PATH}/tgi.log
+        exit 1
+    fi
 
-    docker logs test-comps-intent-server >> ${LOG_PATH}/intent_detection.log
-    docker logs test-comps-intent-tgi-endpoint >> ${LOG_PATH}/tgi-endpoint.log
 }
 
 function stop_docker() {