Performance for mixed precision Resnets?

mit-han-lab / haq

[CVPR 2019, Oral] HAQ: Hardware-Aware Automated Quantization with Mixed Precision

https://hanlab.mit.edu/projects/haq/

MIT License

365 stars 85 forks source link

Performance for mixed precision Resnets? #20

Open kentaroy47 opened 3 years ago

kentaroy47 commented 3 years ago

Thanks for sharing a great work!

I was interested in the results of deep compression vs HAQ in the paper: in case of resnets, mixed precision quantization did not bring much benefits. I was wondering if mixed precision quantization is effective in resnet architecture or not. Did you compare the results of HAQ vs PACT with resnets? It would be great if you can share!