This is a template for general deep learning projects.
.
├─ data/
│ ├─ raw/ <- raw datasets, e.g. downloaded archives
│ ├─ interim/ <- extracted datasets in its raw folder structure
│ └─ processed/ <- processed, e.g. converted and filtered,
│ data ready for training
├─ docs/ <- project documentation
├─ notebooks/ <- Jupyter notebooks used for experiments
├─ models/ <- pre-trained weights and frozen models
├─ src/
│ ├─ data/ <- data loader and processing
│ ├─ models/ <- model implementations
│ ├─ utils/ <- utility functions and classes
│ └─ ... <- files for training, evaluation and inference etc.
├─ training/ <- run configurations and saved checkpoints
│ └─ run_*/ created by src/train.py
├─ LICENSE.md
└─ README.md
Explain here, what the first steps are to get started. This can be stuff like downloading datasets and extracting them into the data folder.
List here dependencies of the project, e.g.
The project was compiled using the following packages:
- Matplotlib 2.2.2 (Information)
- NumPy 1.14.5 (Information)
Give here information about the type of license the project is published with, e.g.
All python code in this repository is available under an MIT license.