rusty1s / deep-learning-cheatsheet

Deep Learning Cheatsheet

General | Graphs

Introduction

• Machine Learning Glossary
• Essential Machine Learning Cheatsheets
• Neural Networks and Deep Learning [Free Online Book]
• Free Deep Learning Book [MIT Press]
• Andrew Ng's machine learning course at Coursera [Material]
• Deep Learning by Google
• Deep Learning in Neural Networks: An Overview
• How To Become A Machine Learning Engineer: Learning Path
• Awesome Deep Vision
• A Guide to Deep Learning
• Deep Learning Weekly [Weekly Newsletter Subscription]
• A Year of Artificial Intelligence [Blog]
• Introduction to Convolutional Neural Networks
• Applied Deep Learning
• Machine Learning is Fun! [Medium Series]
• Machine Learning for Humans [Medium Series]
• My Neural network ins't working! What should I do?
• The math of neural networks
• Everything you need to know about Neural Networks

Python

• Learn Python the hard way
• Python Data Science Handbook
• Styleguide [PEP8]
• TensorFlow [Open Source Library]
• PyCUDA
• Cython
  • Pure Python Mode

TensorFlow

• Official
• Adding a New Op [Official]
• TensorFlow in a Nutshell
• Source Code Examples
• Effictive TensorFlow [TensorFlow tutorials and best practices]

PyTorch

• Official
• PyTorch Examples

Conferences

• ICLR 2018

Architectures

• Very Deep Convolutional Networks for Large-Scale Image Recognition [VGG]
• SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and 0.5MB model size
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• Dynamically Expandable Neural Networks

AutoEncoder and GANs

• Generative Adversarial Networks (GANs): Engine and Applications
• GANs are Broken in More than One Way: The Numerics of GANs
• Towards data set augmentation with GANs
• How does the unpooling and deconvolution work in DeConvNet
• What are deconvoltional layers?
• Visualizing and Understanding Convolutional Networks [Original Unpooling Paper]

Optimizer

• A method for stochastic optimization [Adam Optimizer]

Overfitting

• Dropout: A Simple Way to Prevent Neural Networks from Overfitting
• A Simple Weight Decay Can Improve Generalization
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
  • Batch Normalization - What the hey? [Blog]

Recurrent Neural Networks

• The Unreasonable Effectiveness of Recurrent Neural Networks

Datasets

• Overview

Images

• MNIST
• Fashion MNIST
• CIFAR-10/100
• STL-10
• SVHN
• ImageNet
  • ILSVRC2014
• Pascal VOC
• Animals with Attributes

Meshes

• MPI FAUST Dataset
• Tosca

Classification

• Image Classification datasets results