facebookresearch/maskrcnn-benchmark

Half/mixed-precision is not faster

Open

#1062 aperta il 27 ago 2019

Vedi su GitHub
 (3 commenti) (0 reazioni) (0 assegnatari)Python (2574 fork)batch import
help wanted

Metriche repository

Star
 (9161 star)
Metriche merge PR
 (Nessuna PR mergiata in 30 g)

Descrizione

❓ Questions and Help

As written in https://github.com/facebookresearch/maskrcnn-benchmark/issues/807#issuecomment-500112612 , DTYPE "float16" does not make training faster( and takes the same amount of memory) when tested with e2e_mask_rcnn_R_50_FPN_1x.yaml. Is this normal or am I doing something wrong? I also want to know if I can make inference faster by FP16(I tried model.half() in test_net, but doesn't work).
My environment is RTX Titan, cudnn 7.6, cuda10.0, driver version Driver Version: 418.67, pytorch-1.2.0, ubuntu 18.04.

Guida contributor