Chen Shangyu issues

Repositories
Issues
Comments

Results 3 issues of


                                            Chen Shangyu

[compressor] Codes for post-training quantization & pruning

# Quantization - 4-bit quantization: https://github.com/submission2019/cnn-quantization # Pruning - You may try pruning by absoluted value: Learning both Weights and Connections for Efficient Neural Networks

How to change 5 bits to 4,3, 2 bits?

Hi, I want to change using 5 bits to 4,3,2 bits. How can I do so in the code? Best regards

Reproduced ResNet18 CIFAR10 result is 10% lower than reported

Hi @czhu95 , Thanks for providing the codes! Recently I use your codes to ternarize a ResNet18 using CIFAR10. Firstly I use tensorpack to train a ResNet18 to validation error...