666DZY666/micronet
micronet, a model compression and deploy lib. compression: 1、quantization: quantization-aware-training(QAT), High-Bit(>2b)(DoReFa/Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference)、Low-Bit(≤2b)/Ternary and Binary(TWN/BNN/XNOR-Net); post-training-quantization(PTQ), 8-bit(tensorrt); 2、 pruning: normal、regular and group convolutional channel pruning; 3、 group convolution structure; 4、batch-normalization fuse for quantization. deploy: tensorrt, fp32/fp16/int8(ptq-calibration)、op-adapt(upsample)、dynamic_shape
Topics
Explore related topics
Jump into the topic listings this repository belongs to.
Join the conversation
Reviews · Questions · Posts
Share what you know about micronet — write a review from your real experience, ask an implementation question, or publish a post about how you use it.
Share your experience
Write or update your review
Explain what worked, what broke down, and what another team should know before adopting micronet.
Project Q&A
Questions and answers
Browse implementation threads tied directly to 666DZY666/micronet. Each question links through to the full answer page.
Be the first to ask how teams run micronet in production. Every question you post becomes a durable, searchable answer page other developers can find.
Ask the first questionRelated posts
Posts tagged with the same topics
These posts come from the same topic surface as this repo, so readers can move from project evaluation into practical writeups and migration notes without leaving context.
Share how your team uses micronet — a migration note, an architecture writeup, or a comparison. Your post reaches everyone browsing these same topics.
Write the first post