移植到目标检测模型效果不佳

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

移植到目标检测模型效果不佳 #46

Open lovepan1 opened 3 years ago

lovepan1 commented 3 years ago

大佬好，我将dorefanet移植到基于resnet的backbone的检测网络，在voc数据集上测试，原始fp32精度大概在0.80左右，移植后进行量化训练，我量化了整个backbone，但是保留了head与fpn结构的fp卷积，精度大概在0.13左右，不甚理想，请问有什么tricks吗

charles-str commented 3 years ago

你好，可以一起讨论下压缩方法移到自己的模型中问题吗？

666DZY666 commented 3 years ago

1、用16-bit试试；2、用IAO试试。输入不要量化。