Linear Fitting with Custom CUDA Kernel

This project demonstrates how to implement a simple linear regression model using a custom CUDA kernel and integrate it with PyTorch's autograd system via torch.autograd.Function.

Installation

This project uses pyproject.toml to manage dependencies.

Clone the repository:

git clone git@github.com:sh1027/pytorch-custom-cuda-tutorial.git
cd pytorch-custom-cuda-tutorial

Install Python dependencies using uv:

# Make sure you have uv installed. Check (https://docs.astral.sh/uv/#installation) for the details
uv sync

Or, with pip:

pip install .

Additional system dependencies

• CUDA Toolkit
• ffmpeg (for saving videos)