PyTorch Project Structure
A scalable and maintainable folder structure for PyTorch applications
File Structure
- src
- train.py
- model.py
- dataset.py
- utils.py
- data
- raw
- processed
- models
- checkpoints
- requirements.txt
- README.md
Directory Structure Explanation
src/
Contains the main Python source code for the PyTorch project.
src/train.py
Main training script that defines the training loop and model training process.
src/model.py
Defines the neural network architecture using PyTorch modules.
src/dataset.py
Custom PyTorch Dataset class for loading and preprocessing data.
src/utils.py
Utility functions for data processing and helper methods.
data/
Contains datasets organized by processing stage.
models/checkpoints/
Saved model checkpoints during training for recovery and evaluation.