songhan
Song Han is an associate professor at MIT EECS and distinguished scientist at NVIDIA. His research interest is efficient AI computing.
MIT, NVIDIA
Pinned Repositories
gan-compression
[CVPR 2020] GAN Compression: Efficient Architectures for Interactive Conditional GANs
once-for-all
[ICLR 2020] Once for All: Train One Network and Specialize it for Efficient Deployment
proxylessnas
[ICLR 2019] ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
pvcnn
[NeurIPS 2019, Spotlight] Point-Voxel CNN for Efficient 3D Deep Learning
temporal-shift-module
[ICCV 2019] TSM: Temporal Shift Module for Efficient Video Understanding
tinyml
Deep-Compression-AlexNet
Deep Compression on AlexNet
DSD
DSD model zoo. Better accuracy models from DSD training on Imagenet with same model architecture
SqueezeNet-Deep-Compression
SqueezeNet-Residual
residual-SqueezeNet
songhan's Repositories
songhan/Deep-Compression-AlexNet
Deep Compression on AlexNet
songhan/SqueezeNet-Deep-Compression
songhan/SqueezeNet-Residual
residual-SqueezeNet
songhan/DSD
DSD model zoo. Better accuracy models from DSD training on Imagenet with same model architecture
songhan/SqueezeNet-DSD-Training
songhan/SqueezeNet-Generator
SqueezeNet Generator
songhan/convnet-benchmarks
Easy benchmarking of all public open-source implementations of convnets
songhan/SqueezeNet
SqueezeNet: AlexNet-level accuracy with 50x fewer parameters
songhan/dotfiles
my dotfiles..
songhan/dotvim
Over 1200+ lines of vimrc
songhan/haDNN
Proof-of-Concept CNN in Halide
songhan/Halide
a language for image processing and computational photography
songhan/neural-style
Torch implementation of neural style algorithm