facebookresearch/maskrcnn-benchmark

Inference time increases linearly with batch size

Open

#294 aperta il 20 dic 2018

Vedi su GitHub
 (8 commenti) (2 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

I tested the batch inference,this is result batch size time 1 0.22 2 0.42 4 0.84 6 1.2 8 crash OOM Why big batchsize doesn't save time?

Guida contributor