CodeTrans¶

Helm chart for deploying CodeTrans service.

CodeTrans depends on LLM microservice, refer to llm-uservice for more config details.

Installing the Chart¶

To install the chart, run the following:

cd GenAIInfra/helm-charts/
./update_dependency.sh
helm dependency update codetrans
export HFTOKEN="insert-your-huggingface-token-here"
export MODELDIR="/mnt/opea-models"
export MODELNAME="HuggingFaceH4/mistral-7b-grok"
helm install codetrans codetrans --set global.HUGGINGFACEHUB_API_TOKEN=${HFTOKEN} --set global.modelUseHostPath=${MODELDIR} --set tgi.LLM_MODEL_ID=${MODELNAME}
# To use Gaudi device
# helm install codetrans codetrans --set global.HUGGINGFACEHUB_API_TOKEN=${HFTOKEN} --values codetrans/gaudi-values.yaml

Verify¶

To verify the installation, run the command kubectl get pod to make sure all pods are running.

Curl command and UI are the two options that can be leveraged to verify the result.

Verify the workload through curl command¶

Then run the command kubectl port-forward svc/codetrans 7777:7777 to expose the CodeTrans service for access.

Open another terminal and run the following command to verify the service if working:

curl http://localhost:7777/v1/codetrans \
    -H 'Content-Type: application/json' \
    -d '{"language_from": "Golang","language_to": "Python","source_code": "package main\n\nimport \"fmt\"\nfunc main() {\n    fmt.Println(\"Hello, World!\");\n}"}'

Verify the workload through UI¶

The UI has already been installed via the Helm chart. To access it, use the external IP of one your Kubernetes node along with the NGINX port. You can find the NGINX port using the following command:

export port=$(kubectl get service codetrans-nginx --output='jsonpath={.spec.ports[0].nodePort}')
echo $port

Open a browser to access http://<k8s-node-ip-address>:${port} to play with the ChatQnA workload.

Values¶

Key	Type	Default	Description
image.repository	string	`"opea/codetrans"`
service.port	string	`"7777"`
tgi.LLM_MODEL_ID	string	`"HuggingFaceH4/mistral-7b-grok"`	Models id from https://huggingface.co/, or predownloaded model directory