pytorch_net

This repository provides classes for

Efficient construction of neural networks with PyTorch, including multilayer perceptron (MLP), LSTM, CNN, WideResNet, model ensembles, etc.
Easy manipulation of networks, including addition and removal of layers and neurons, training of networks, simplification of networks, loading and saving models as a dictionary. See tutorial.ipynb for a simple demonstration.

Requirements:

Name	Name	Last commit message	Last commit date
Latest commit tailintalent Update kernel_condition Mar 24, 2023 ff2e5db · Mar 24, 2023 History 495 Commits
transformer @ 4f0db4b	transformer @ 4f0db4b	Minor	Jul 31, 2022
.gitignore	.gitignore	Initial commit	Jun 19, 2018
.gitmodules	.gitmodules	Add transformer	Jul 22, 2022
LICENSE	LICENSE	Initial commit	Jun 19, 2018
README.md	README.md	Update README	Mar 13, 2023
Tutorial.ipynb	Tutorial.ipynb	Improve to_symbolic function to deal with activations.	Dec 8, 2019
Tutorial.py	Tutorial.py	Improve to_symbolic function to deal with activations.	Dec 8, 2019
__init__	__init__	Add MLP, LSTM and Simple_Layer modules, and supporting functions	Jun 19, 2018
logger.py	logger.py	Change default scheduler lambda_lr	Jan 31, 2020
mi.py	mi.py	Add mi estimators	May 1, 2021
mi_test.ipynb	mi_test.ipynb	Add mi estimators	May 1, 2021
modules.ipynb	modules.ipynb	Add __repr__ for Simple_Layer	May 19, 2020
modules.py	modules.py	Add __repr__ for Simple_Layer	May 19, 2020
net.ipynb	net.ipynb	Add dostring	Sep 16, 2022
net.py	net.py	Add dostring	Sep 16, 2022
requirements.txt	requirements.txt	Update requirements	Oct 18, 2019
swd.py	swd.py	Able to apply to different n_channels.	Sep 23, 2021
util.py	util.py	Update kernel_condition	Mar 24, 2023