This repository can be used as a starter for contests on kaggle. It has placeholders for the most commonly encountered functionalities which can be implemented and extended as seen fit.
- data
- file_list.md
- src
- utils.py
- train.py
- valid.py
- models.py
- config.py
- create_folds.py
- saved_models
- model_list.md
- notebooks
- check_logging.ipynb
- eda.ipynb
- references
- list.md
- requirements.txt
- The
data
folder contains the input files, images, etc for the contest - The
src
folder contains all source code for the project except any jupyter notebooks - The
saved_models
folder contains weights and checkpoints for models we would like to use for inference - The
notebooks
folder contains jupyter notebooks used for quick debugging, exploratory data analysis, etc - The
references
folder contains a list of reference papers, articles, etc relevant to the contest - The
requirements.txt
is an optional file that can contain a list of requirements to run the code in the repository