Skip to content
Learn PyTorch mainly by "Deep Learning with PyTorch"
Jupyter Notebook Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.github
.vscode
data
images
.gitignore
GAN.py
GPU-16.ipynb
LICENSE
chapter2-Building Blocks Of Neural Networks.ipynb
chapter3-Image Classification.ipynb
chapter6-Sequence Data and Text.ipynb
chapter7-DCGAN.ipynb
chapter7-Language Modeling.ipynb
chapter7-Style transfer.ipynb
ensemble.py
image-classify.py
only-fc.py
preconvolute.py
readme.md
utility.py

readme.md

Overview

Deep Learning with PyTorch is a very "good" book to teach oneself PyTorch. Because it's old, incomplete, even with error, and those force you thinking, exploring, therefore learning.

But reproducibility is also very important, sometimes we need code ready for use. That's why I create this repo.

to do

open question

  • why only-fc.py is better than image-classify.py --only-fc?
  • For resnet, training use dropout, why result is still reproduciable.

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported License

You can’t perform that action at this time.