How to calculate FLOP? - Githubissues

HRNet / HRNet-Object-Detection

Object detection with multi-level representations generated from deep high-resolution representation learning (HRNetV2h). This is an official implementation for our TPAMI paper "Deep High-Resolution Representation Learning for Visual Recognition". https://arxiv.org/abs/1908.07919

Apache License 2.0

644 stars 97 forks source link

How to calculate FLOP? #21

Closed ElegantLin closed 5 years ago

ElegantLin commented 5 years ago

How do you calculate your FLOP?

I need the data.

Thanks a lot.

wondervictor commented 5 years ago

We set the image size to 800*1200 and only calculate the FLOPs statistics of Convolutional layers and Batch Normalization layers. For R-CNN, we set the number of ROIs to 512.

wondervictor commented 5 years ago

No more response, closed