Convolutional Neural Networks (CNN)


A convolutional neural network (CNN, or ConvNet) is a class of deep neural networks, most commonly applied to analyzing visual imagery.

Overview

Common Architectures in Convolutional Neural Networks
In this post, I'll discuss commonly used architectures for convolutional networks.
convolutional-neural-networks architectures computer-vision survey
How to Derive Convolution From First Principles
I derive the convolution from first principles and show that it naturally emerges from translational symmetry.
convolutional-neural-networks article

Tutorials

CS231n: Convolutional Neural Networks for Visual Recognition
Deep dive into details of the deep learning architectures with a focus on learning end-to-end models for these tasks, particularly image classification.
convolutional-neural-networks computer-vision deep-learning cs231n
Understanding Convolutional Neural Networks for NLP
More recently we’ve also started to apply CNNs to problems in Natural Language Processing and gotten some interesting results.
convolutional-neural-networks natural-language-processing text-classification tutorial

Libraries

General
PyTorch CNN Trainer
A simple package to fine-tune CNNs from torchvision and Pytorch Image models by Ross Wightman.
torchvision convolutional-neural-networks pytorch training
CNN Explainer
CNN Explainer uses TensorFlow.js, an in-browser GPU-accelerated deep learning library to load the pretrained model for visualization.
convolutional-neural-networks tensorflow-js interactive interpretability
ConvNet Playground
An interactive visualization for exploring Convolutional Neural Networks applied to the task of semantic image search.
interactive convolutional-neural-networks playground data-visualization
Antialiased CNNs
Making Convolutional Networks Shift-Invariant Again.
convolutional-neural-networks shift-invariant paper video
Table of Contents
Share a project
Share something you or the community has made with ML.
Topic experts
Share