search

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
MIT License
2.2k stars 478 forks source link

如何压缩自己的模型？ #56

Open charles-str opened 3 years ago

charles-str commented 3 years ago

大佬，如何将上述方法应用到自己的模型呢？

zhangfeixiang222 commented 3 years ago

应该需要使用他的方法重新构建你的网络，之后再训练

666DZY666 commented 3 years ago

参考，使用_迁移