/ronald_bdl

An experimental Python package for learning Bayesian Neural Network.

Primary LanguagePythonMIT LicenseMIT

ronald_bdl

ronald_bdl is the Python package used to produce the results presented in the technical report Qualitative Analysis of Monte Carlo Dropout.

This package contains pre-defined PyTorch NN models (ronald_bdl.models) and Dataset classes (ronald_bdl.datasets). Please run the following command using pip to install:

pip install git+https://github.com/ronaldseoh/ronald_bdl.git

If you want to modify the code within ronald_bdl, please clone/download the ronald_bdl repo, apply your changes, and install your version using the command pip install .

Please refer to README.md in the bayesian-dl-experiments repo for more details.

This repository was started with the codes transferred from bayesian-dl-experiments. For older commit histories, please check here.

License

ronald_bdl is licensed under MIT license. Please check LICENSE.