model-deployment-workshop

1. Create a new repository if not created already

2. Setup environment

   1. Install asdf using this guide
   2. Create python environment

      > python -m venv ml
      > chmod +x ml/bin/activate
      > source ml/bin/activate
      

   3. Install packages

      pip install -r requirements.txt
      

3. Train model (if mode)l.pkl file is not saved in model/)

   cd tabular
   python model.py
   

4. Run the app locally

   1. Run flask app

      cd tabular
      python api.py
      

   2. Run the streamlit app

      streamlit run ui.py
      

5. Run the app locally using Docker

   1. Install docker(skip if already installed)

      • Steps to install Docker on your system

   2. Create docker image

      sudo docker build -t ml_deployment .
      

   3. Run docker container

      sudo docker run -d -p 8501:8501 -p 5000:5000 ml_deployment