A PyTorch CV Toolkit

How to use Cross Replica / Synchronized Batchnorm in Pytorch

A Re-implementation of Fixed-update Initialization

MXNet Gluon Synchronized Batch Normalization Preview

Rethinking the Smaller-Norm-Less-Informative Assumption in Channel Pruning of Convolution Layers https://arxiv.org/abs/1802.00124

Review materials for the TWiML Study Group. Contains annotated versions of the original Jupyter noteboooks (look for names like *_jcat.ipynb ), slide decks from weekly Zoom meetups, etc.

My solutions for Assignments of CS231n: Convolutional Neural Networks for Visual Recognition

VGG16 architecture with BatchNorm

Stochastic Downsampling for Cost-Adjustable Inference and Improved Regularization in Convolutional Networks

Deep learning projects including applications (face recognition, neural style transfer, autonomous driving, sign language reading, music generation, translation, speech recognition and NLP) and theories (CNNs, RNNs, LSTM, Adam, Dropout, BatchNorm, Xavier/He initialization, hyperparameter tuning, regularization, optimization, Residual Networks). Deep Learning Specialization by Andrew Ng, deeplearning.ai

Synchronized BatchNorm in PyTorch 1.0

MNIST Classification using Neural Network and Back Propagation. Written in Python and depends only on Numpy

remove batchnorm when deploy mxnet model

fuse batch normalization layer and con layer in CNN

This repository contains different implementation of deep learning model using Numpy.

Digit recognition neural network using the MNIST dataset. Features include a full gui, convolution, pooling, momentum, nesterov momentum, RMSProp, batch normalization, and deep networks.

Neural network with momentum and batchnorm (MNIST classification example)

I implemented a classifier using batchnorm provided by tensorflow slim and analyzed the results using tensorboard.

Built CNN model with different no. of layers with added dropout on MNIST data

MXNet implementation of Filter Response Normalization Layer (FRN) published in CVPR2020

Built MLP with ReLU and Adam optimization with 2 layers, 3 layers and 5 layers and observed how it works.

A C# WGAN.