March 15, 2019: for our most updated work on model compression and acceleration, please reference:
Die Hauptfunktionen von songhan/deep-compression-alexnet sind: Efficient Neural Networks, Weight Pruning.
Open-Source-Alternativen zu songhan/deep-compression-alexnet sind unter anderem: tencent/pocketflow — PocketFlow is an integrated toolkit for deep learning model compression, distributed training, and mobile format… boone891214/gap — ICLR 2022 paper "Effective Model Sparsification by Scheduled Grow-and-Prune Methods". Model and test code are… boone891214/sanity-check-lth — Sample code use for NeurIPS 2021 paper: Sanity Checks for Lottery Tickets: Does Your Winning Ticket Really Win the… chrundle/biprop — This method identifies a binary weight or binary weight and activation subnetwork within a randomly initialized… dchiji-ntt/iterand — by Daiki Chijiwa\*, Shin’ya Yamaguchi, Yasutoshi Ida, Kenji Umakoshi, Tomohiro Inoue. allenai/xnor-net.
PocketFlow is an integrated toolkit for deep learning model compression, distributed training, and mobile format optimization. It provides a system for reducing the size and complexity of neural networks to improve inference efficiency, featuring a dedicated engine for knowledge distillation and a mobile model optimizer. The framework differentiates itself through an automated hyperparameter tuning system that uses reinforcement learning and statistical models to determine optimal compression ratios and layer-wise bit allocation. It also includes a distributed training system that utilizes mu
ICLR 2022 paper "Effective Model Sparsification by Scheduled Grow-and-Prune Methods". Model and test code are available for downloading.
Sample code use for NeurIPS 2021 paper: Sanity Checks for Lottery Tickets: Does Your Winning Ticket Really Win the Jackpot?