Machine Learning Engineering (MLE) with a focus on deployment of ML models in production.
FLAN-T5 Small Text Generation with FastAPI
- app.py: model inference
To call the inference API, use the following command:
curl -X 'GET' \
'http://localhost:8080/generate?text=what%20is%20the%20meaning%20of%20life%3F' \
-H 'accept: application/json'
RoBERTa sentiment classification
- inference.py: model inference
To call the inference API, use the following command:
curl -X POST -H "Content-Type: application/JSON" --data '["Deploying ML models is fun!"]' http://0.0.0.0:5000/predict
The docker image is available on docker hub: https://hub.docker.com/repository/docker/vsmolyakov/roberta-sentiment
Random Forest Classifier Docker Deployment
- train.py: classifier training script
- inference.py: model inference
- api_call.py: API endpoint call
Docker commands:
docker build -t iris-rfc .
docker run -p 5000:5000 iris-rfc
The docker image is available on docker hub: https://hub.docker.com/repository/docker/vsmolyakov/iris-rfc