To run the project locally install the requirements
run pip install -r requirements.txt
after installation to save the model to bento store
run python3 save_model.py
This will download and export the model to bento store.
if you want to use a different model, update constants.py with the model details. Make sure model under betofile.yaml and MODEL_TAG are same.
After finishing the above steps, you can run the service locally or via docker.
To run locally
run bentoml serve
server runs at 3000, you can change the port by modifying the port configured in service.py
To create docker
run bentoml build
it will provide the command for dockerizing the service, on running the command docker image will be generated.
There is a test.py file available to test the service. open a new terminal r
run python3 test.py