Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- --------------------------------------------------------------------------------
- Environment Summary
- --------------------------------------------------------------------------------
- PyTorch 1.4.0 compiled w/ CUDA 10.1
- Running with Python 3.7 and
- `pip list` truncated output:
- numpy==1.18.1
- torch==1.4.0
- torchvision==0.5.0
- --------------------------------------------------------------------------------
- cProfile output
- --------------------------------------------------------------------------------
- 1138288 function calls (1108000 primitive calls) in 27.004 seconds
- Ordered by: internal time
- List reduced from 6815 to 15 due to restriction <15>
- ncalls tottime percall cumtime percall filename:lineno(function)
- 98 20.098 0.205 20.098 0.205 {method 'to' of 'torch._C._TensorBase' objects}
- 5 3.306 0.661 3.306 0.661 {built-in method ctc_loss}
- 93 1.078 0.012 1.078 0.012 {method 'cuda' of 'torch._C._TensorBase' objects}
- 28 0.434 0.016 0.434 0.016 {method 'uniform_' of 'torch._C._TensorBase' objects}
- 1545 0.333 0.000 0.333 0.000 {built-in method numpy.fft._pocketfft_internal.execute}
- 160 0.220 0.001 0.707 0.004 /opt/miniconda3/envs/pytorchasr/lib/python3.7/site-packages/librosa/core/spectrum.py:2461(_spectrogram)
- 70 0.119 0.002 0.119 0.002 {built-in method conv1d}
- 160 0.109 0.001 0.487 0.003 /opt/miniconda3/envs/pytorchasr/lib/python3.7/site-packages/librosa/core/spectrum.py:34(stft)
- 1066 0.102 0.000 0.102 0.000 {built-in method marshal.loads}
- 160 0.079 0.000 0.108 0.001 /opt/miniconda3/envs/pytorchasr/lib/python3.7/site-packages/librosa/filters.py:112(mel)
- 160 0.061 0.000 0.064 0.000 /opt/miniconda3/envs/pytorchasr/lib/python3.7/site-packages/librosa/core/spectrum.py:1507(power_to_db)
- 28982 0.031 0.000 0.032 0.000 {built-in method builtins.getattr}
- 160 0.028 0.000 0.028 0.000 {built-in method scipy.fft._pocketfft.pypocketfft.dct}
- 9777/3967 0.027 0.000 0.434 0.000 {built-in method numpy.core._multiarray_umath.implement_array_function}
- 2912/2898 0.027 0.000 0.085 0.000 {built-in method builtins.__build_class__}
- --------------------------------------------------------------------------------
- autograd profiler output (CPU mode)
- --------------------------------------------------------------------------------
- top 15 events sorted by cpu_time_total
- ------------ --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- -----------------------------------
- Name Self CPU total % Self CPU total CPU total % CPU total CPU time avg CUDA total % CUDA total CUDA time avg Number of Calls Input Shapes
- ------------ --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- -----------------------------------
- to 16.56% 5.489s 16.56% 5.489s 5.489s NaN 0.000us 0.000us 1 []
- to 16.29% 5.399s 16.29% 5.399s 5.399s NaN 0.000us 0.000us 1 []
- to 16.16% 5.355s 16.16% 5.355s 5.355s NaN 0.000us 0.000us 1 []
- to 15.23% 5.048s 15.23% 5.048s 5.048s NaN 0.000us 0.000us 1 []
- to 14.94% 4.952s 14.94% 4.952s 4.952s NaN 0.000us 0.000us 1 []
- ctc_loss 2.11% 698.115ms 2.11% 698.115ms 698.115ms NaN 0.000us 0.000us 1 []
- to 2.10% 696.654ms 2.10% 696.654ms 696.654ms NaN 0.000us 0.000us 1 []
- ctc_loss 2.10% 696.523ms 2.10% 696.523ms 696.523ms NaN 0.000us 0.000us 1 []
- ctc_loss 2.10% 695.301ms 2.10% 695.301ms 695.301ms NaN 0.000us 0.000us 1 []
- to 2.10% 695.000ms 2.10% 695.000ms 695.000ms NaN 0.000us 0.000us 1 []
- ctc_loss 2.09% 694.288ms 2.09% 694.288ms 694.288ms NaN 0.000us 0.000us 1 []
- to 2.09% 693.801ms 2.09% 693.801ms 693.801ms NaN 0.000us 0.000us 1 []
- to 2.09% 692.779ms 2.09% 692.779ms 692.779ms NaN 0.000us 0.000us 1 []
- ctc_loss 2.02% 668.966ms 2.02% 668.966ms 668.966ms NaN 0.000us 0.000us 1 []
- to 2.01% 667.492ms 2.01% 667.492ms 667.492ms NaN 0.000us 0.000us 1 []
- ------------ --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- -----------------------------------
- Self CPU time total: 33.142s
- CUDA time total: 0.000us
- --------------------------------------------------------------------------------
- autograd profiler output (CUDA mode)
- --------------------------------------------------------------------------------
- top 15 events sorted by cpu_time_total
- Because the autograd profiler uses the CUDA event API,
- the CUDA time column reports approximately max(cuda_time, cpu_time).
- Please ignore this output if your code does not use CUDA.
- ------------ --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- -----------------------------------
- Name Self CPU total % Self CPU total CPU total % CPU total CPU time avg CUDA total % CUDA total CUDA time avg Number of Calls Input Shapes
- ------------ --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- -----------------------------------
- addcmul_ 18.41% 4.543s 18.41% 4.543s 4.543s 0.03% 2.000us 2.000us 1 []
- addcmul_ 18.06% 4.455s 18.06% 4.455s 4.455s 0.03% 2.000us 2.000us 1 []
- addcmul_ 17.92% 4.423s 17.92% 4.423s 4.423s 0.05% 4.000us 4.000us 1 []
- addcmul_ 16.88% 4.164s 16.88% 4.164s 4.164s 0.03% 2.000us 2.000us 1 []
- ctc_loss 2.83% 697.706ms 2.83% 697.706ms 697.706ms 20.85% 1.577ms 1.577ms 1 []
- to 2.82% 696.173ms 2.82% 696.173ms 696.173ms 1.11% 84.000us 84.000us 1 []
- ctc_loss 2.81% 693.902ms 2.81% 693.902ms 693.902ms 21.00% 1.588ms 1.588ms 1 []
- ctc_loss 2.81% 693.869ms 2.81% 693.869ms 693.869ms 17.70% 1.339ms 1.339ms 1 []
- to 2.81% 692.557ms 2.81% 692.557ms 692.557ms 0.53% 40.000us 40.000us 1 []
- to 2.81% 692.366ms 2.81% 692.366ms 692.366ms 1.14% 86.000us 86.000us 1 []
- ctc_loss 2.73% 672.559ms 2.73% 672.559ms 672.559ms 17.27% 1.306ms 1.306ms 1 []
- to 2.72% 671.276ms 2.72% 671.276ms 671.276ms 0.50% 38.000us 38.000us 1 []
- ctc_loss 2.71% 669.494ms 2.71% 669.494ms 669.494ms 19.23% 1.454ms 1.454ms 1 []
- to 2.71% 668.190ms 2.71% 668.190ms 668.190ms 0.50% 38.000us 38.000us 1 []
- add_ 0.97% 239.802ms 0.97% 239.802ms 239.802ms 0.03% 2.000us 2.000us 1 []
- ------------ --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- --------------- -----------------------------------
- Self CPU time total: 24.673s
- CUDA time total: 7.561ms
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement