Inspired by Stanford CS230 blog post, I created this repository to use as a template for my ML projects. util.py will be updated with helper functions through time.

Main libraries:

• PyTorch: as the main ML framework
• Comet.ml: tracking code, logging experiments
• OmegaConf: for managing configuration files

Installation

First create a virtual env for the project.

python3 -m venv .venv
source .venv/bin/activate

Then install the latest version of PyTorch from the official site. Finally, run the following:

pip install -r requirements

To set up Comet.Ml follow the official documentations.

What's new

• (July 3, 2021) Added run.sh bash file to easily run different parts of the code.
• (July 3, 2021) Added UUID generation for sample names.
• (July 3, 2021) Added MNIST as an example (not SOTA).

Run

To run different parts, just edit run.sh file and run it using . calling.