Torchcutter
A logical, reasonably standardized, but flexible framework for training, evaluating, and deploying PyTorch models.
Project Organization
├── LICENSE
├── Makefile <- Makefile with commands like `make data`, `make train` or 'make deploy', ... etc
├── README.md <- The top-level README for developers using this project.
├── data
│ ├── external <- Data from third party sources.
│ ├── interim <- Intermediate data that has been transformed.
│ ├── processed <- The final, canonical data sets for modeling.
│ └── raw <- The original, immutable data dump.
│
├── docs <- torchcutter documentation
│
├── output <- Trained and serialized models, model predictions, or model summaries
│ └── figures <- Generated model evaluation figures
│ └── models <- Saves checkpoints
│
├── notebooks <- Jupyter notebooks. Naming convention is a number (for ordering),
│ the creator's initials, and a short `-` delimited description, e.g.
│ `1.0-jqp-initial-data-exploration`.
│
├── references <- Data dictionaries, manuals, and all other explanatory materials.
│
├── reports <- Generated analysis as HTML, PDF, LaTeX, etc.
│ └── figures <- Generated graphics and figures to be used in reporting
│
├── requirements.txt <- The requirements file for reproducing the analysis environment, e.g.
│ generated with `pip freeze > requirements.txt`
│
├── setup.py <- makes project pip installable (pip install -e .) so src can be imported
├── src <- Source code for use in this project.
│ ├── __init__.py <- Makes src a Python module
│ │
│ ├── data <- data helpers; classes and functions
│ │ └── dataset.py <- Custom PyTorch dataset
│ │ └── get_data.py <- Get data wrapped into dataloader ready to be used to train model
│ │
│ ├── models <- Model architecture to train
│ │ └── model.py <- Model architecture in PyTorch
│ │
│ ├── tester <- Testing and make inference using the pretrained checkpoints
│ │ ├── base.py <- Abstract class for testing model
│ │ └── tester.py <- Class implementing routines to make inference with a checkpoint
│ │
│ ├── trainer <- Testing and make inference using the pretrained checkpoints
│ │ ├── base.py <- Abstract class for training model
│ │ └── tester.py <- Class implementing routines to train the model
│ │
│ ├── utils <- utility classes and function
│ │ ├─── io <- Module for input/output operations
│ │ │ ├─── read.py <- Read from disk functions utility, like Read configuration file, ... etc
│ │ │ ├─── write.py <- write to disk functions utility, liek write to csv file, ... etc
│
└── tox.ini <- tox file with settings for running tox; see tox.readthedocs.io