Skip to content

Instantly share code, notes, and snippets.

@lanpa

lanpa/autograd_profiler.txt

Created May 21, 2018 08:51
Show Gist options
  • Save lanpa/3b8b761b65568be1935784cb3b814fc8 to your computer and use it in GitHub Desktop.
Save lanpa/3b8b761b65568be1935784cb3b814fc8 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
autograd_profiler.txt
--------------------------- --------------- --------------- --------------- --------------- ---------------
Name CPU time CUDA time Calls CPU total CUDA total
--------------------------- --------------- --------------- --------------- --------------- ---------------
conv2d 462.645us 0.000us 1 462.645us 0.000us
convolution 461.316us 0.000us 1 461.316us 0.000us
_convolution 459.809us 0.000us 1 459.809us 0.000us
tensor 2.967us 0.000us 1 2.967us 0.000us
_convolution_nogroup 445.746us 0.000us 1 445.746us 0.000us
thnn_conv2d 441.713us 0.000us 1 441.713us 0.000us
thnn_conv2d_forward 439.451us 0.000us 1 439.451us 0.000us
max_pool2d 278.590us 0.000us 1 278.590us 0.000us
max_pool2d_forward 277.156us 0.000us 1 277.156us 0.000us
relu 23.545us 0.000us 1 23.545us 0.000us
neg 21.381us 0.000us 1 21.381us 0.000us
relu 71.921us 0.000us 1 71.921us 0.000us
add 57.257us 0.000us 1 57.257us 0.000us
conv2d 442.940us 0.000us 1 442.940us 0.000us
convolution 442.042us 0.000us 1 442.042us 0.000us
_convolution 441.220us 0.000us 1 441.220us 0.000us
tensor 1.020us 0.000us 1 1.020us 0.000us
_convolution_nogroup 434.149us 0.000us 1 434.149us 0.000us
thnn_conv2d 432.182us 0.000us 1 432.182us 0.000us
thnn_conv2d_forward 431.057us 0.000us 1 431.057us 0.000us
FeatureDropout 259.280us 0.000us 1 259.280us 0.000us
clone 12.675us 0.000us 1 12.675us 0.000us
tensor 1.118us 0.000us 1 1.118us 0.000us
bernoulli_ 56.333us 0.000us 1 56.333us 0.000us
tensor 2.872us 0.000us 1 2.872us 0.000us
fill_ 1.653us 0.000us 1 1.653us 0.000us
expand 4.087us 0.000us 1 4.087us 0.000us
bernoulli 20.650us 0.000us 1 20.650us 0.000us
div_ 4.065us 0.000us 1 4.065us 0.000us
expand 3.581us 0.000us 1 3.581us 0.000us
mul_ 37.522us 0.000us 1 37.522us 0.000us
max_pool2d 56.139us 0.000us 1 56.139us 0.000us
max_pool2d_forward 55.545us 0.000us 1 55.545us 0.000us
relu 7.731us 0.000us 1 7.731us 0.000us
batch_norm 82.295us 0.000us 1 82.295us 0.000us
thnn_batch_norm 79.020us 0.000us 1 79.020us 0.000us
thnn_batch_norm_forward 72.651us 0.000us 1 72.651us 0.000us
view 9.270us 0.000us 1 9.270us 0.000us
unsigned short 3.672us 0.000us 1 3.672us 0.000us
expand 4.277us 0.000us 1 4.277us 0.000us
addmm 30.915us 0.000us 1 30.915us 0.000us
relu 3.959us 0.000us 1 3.959us 0.000us
Dropout 66.604us 0.000us 1 66.604us 0.000us
clone 2.316us 0.000us 1 2.316us 0.000us
tensor 0.885us 0.000us 1 0.885us 0.000us
bernoulli_ 28.035us 0.000us 1 28.035us 0.000us
tensor 1.843us 0.000us 1 1.843us 0.000us
fill_ 0.768us 0.000us 1 0.768us 0.000us
expand 2.797us 0.000us 1 2.797us 0.000us
bernoulli 15.955us 0.000us 1 15.955us 0.000us
div_ 1.108us 0.000us 1 1.108us 0.000us
expand 2.564us 0.000us 1 2.564us 0.000us
mul_ 1.135us 0.000us 1 1.135us 0.000us
unsigned short 3.572us 0.000us 1 3.572us 0.000us
expand 3.701us 0.000us 1 3.701us 0.000us
addmm 10.720us 0.000us 1 10.720us 0.000us
softmax 7.324us 0.000us 1 7.324us 0.000us
softmax_forward 6.630us 0.000us 1 6.630us 0.000us
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment