/pytorch-CI-Docker-Github-actions

ML project which is containerized using Docker, Trained with Pytorch & Tensorboard. Additional Github Actions can be added to the workflow file for pushing the Image to ECR/ GCR /Dockerhub.

Primary LanguagePython

For Local Setup

Install requirements

Step 1: Create a new Anaconda/Miniconda environment

conda create -n <choose a name> -f conda.yaml

Step 2: Install packages in an existing environment using pip

pip install -r requirements.txt

Step:3 Application entry point

python main.py

Step:4 Tensorboard

To start Tensorboard:

tensorboard --logdir runs

The output will be something like:

TensorBoard 2.6.0 at http://localhost:6006/ (Press CTRL+C to quit)

Follow the instructions printed to the terminal to view Tensorboard in a browser.

For Building Docker Image

  • Clone the repo (optional)
git clone https://github.com/rakesh-racharla/pytorch-CI-Docker-Github-actions.git
  • Make changes if needed and commit.
  • Every push will trigger a new build.
  • Check your workflow under "Actions" Tab.