Userful pytorch toolkit for training models, replacing boilerplate code. Provides functions for training, modeling and evaluating models. Also provides several architecture implementations.
pip install my-pytorch-kitClone this repo and run pip install ..
Then, you can import the module my_pytorch_kit.
This package revolves around the BaseModel, Trainer and Evaluator classes, which are extended to model, train and evaluate a model respectively.
Furthermore, this package provides architecture implementations and modelling utilities. Currently implemented architectures include:
Lastly the Tuner class provides hyperparameter tuning using grid, random and random dynamic search.
For an initial guide, see the examples/mnist/classifier/example.py file.
Clone this repo and run pip install -e .[dev].
Run pytest in the root directory to run tests.