search

chengyangfu / caffe

Caffe: a fast open framework for deep learning.
http://caffe.berkeleyvision.org/
Other
169 stars 47 forks source link

how to measure fps? #22

Open foralliance opened 6 years ago

foralliance commented 6 years ago

There seem to be two ways of calculating FPS.

  1. via the 'time' model of caffe with batch_size = 1： build/tools/caffe time -model=models/VGGNet/VOC0712/refinedet_vgg16_320x320/deploy.prototxt -gpu=0

  2. get the time interval and estimate the FPS. for example, got the following lines after I run examples/ssd/score_ssd_pascal.py:

    I0713 20:50:25.756376 117599 net.cpp:684] Ignoring source layer mbox_loss
I0713 20:51:32.472930 117599 solver.cpp:531] Test net output #0: detection_eval = 0.723217
I0713 20:51:32.473057 117599 solver.cpp:325] Optimization Done.

    The total time spent on evaluating 4952 test images is about 67 seconds, and thus FPS = 4952 / 67 = 74.

Which method should be used to calculate FPS? Is there any other way?

Angzz commented 6 years ago

I think the dssd is largely memory inefficient, so I think the FPS should be very low.