pytorch

5095332a - Minor cleanup of interpolate microbenchmark

Minor cleanup of interpolate microbenchmark Summary: Minor cleanup, addresses comments from https://www.internalfb.com/diff/D26780116 (https://github.com/pytorch/pytorch/commit/1559fa6a5c10705d5b1380ad782f3f602e263acc) Test Plan: ``` ➜ vision buck run //caffe2/benchmarks/operator_benchmark/pt:interpolate_test -- --tag_filter short Parsing buck files: finished in 0.6 sec Building: finished in 6.2 sec (100%) 10951/10951 jobs, 0 updated Total time: 6.9 sec /data/users/nicolashug/fbsource/fbcode/buck-out/dev/gen/caffe2/benchmarks/operator_benchmark/pt/interpolate_test#link-tree/torch/utils/cpp_extension.py:3: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses import imp # ---------------------------------------- # PyTorch/Caffe2 Operator Micro-benchmarks # ---------------------------------------- # Tag : short # Benchmarking PyTorch: interpolate # Mode: Eager # Name: interpolate_input_size(1,3,60,40)_output_size(24,24)_channels_lastTrue_modenearest # Input: input_size: (1, 3, 60, 40), output_size: (24, 24), channels_last: True, mode: nearest Forward Execution Time (us) : 1346.156 # Benchmarking PyTorch: interpolate # Mode: Eager # Name: interpolate_input_size(1,3,60,40)_output_size(24,24)_channels_lastTrue_modelinear # Input: input_size: (1, 3, 60, 40), output_size: (24, 24), channels_last: True, mode: linear Forward Execution Time (us) : 1283.784 # Benchmarking PyTorch: interpolate # Mode: Eager # Name: interpolate_input_size(1,3,60,40)_output_size(24,24)_channels_lastTrue_modebicubic # Input: input_size: (1, 3, 60, 40), output_size: (24, 24), channels_last: True, mode: bicubic Forward Execution Time (us) : 4769.578 # Benchmarking PyTorch: interpolate # Mode: Eager # Name: interpolate_input_size(1,3,60,40)_output_size(24,24)_channels_lastFalse_modenearest # Input: input_size: (1, 3, 60, 40), output_size: (24, 24), channels_last: False, mode: nearest Forward Execution Time (us) : 982.910 # Benchmarking PyTorch: interpolate # Mode: Eager # Name: interpolate_input_size(1,3,60,40)_output_size(24,24)_channels_lastFalse_modelinear # Input: input_size: (1, 3, 60, 40), output_size: (24, 24), channels_last: False, mode: linear Forward Execution Time (us) : 1182.191 # Benchmarking PyTorch: interpolate # Mode: Eager # Name: interpolate_input_size(1,3,60,40)_output_size(24,24)_channels_lastFalse_modebicubic # Input: input_size: (1, 3, 60, 40), output_size: (24, 24), channels_last: False, mode: bicubic Forward Execution Time (us) : 3545.873 # Benchmarking PyTorch: interpolate # Mode: Eager # Name: interpolate_input_size(1,3,600,400)_output_size(240,240)_channels_lastTrue_modenearest # Input: input_size: (1, 3, 600, 400), output_size: (240, 240), channels_last: True, mode: nearest Forward Execution Time (us) : 34373.955 # Benchmarking PyTorch: interpolate # Mode: Eager # Name: interpolate_input_size(1,3,600,400)_output_size(240,240)_channels_lastTrue_modelinear # Input: input_size: (1, 3, 600, 400), output_size: (240, 240), channels_last: True, mode: linear Forward Execution Time (us) : 42248.109 # Benchmarking PyTorch: interpolate # Mode: Eager # Name: interpolate_input_size(1,3,600,400)_output_size(240,240)_channels_lastTrue_modebicubic # Input: input_size: (1, 3, 600, 400), output_size: (240, 240), channels_last: True, mode: bicubic Forward Execution Time (us) : 405944.286 ... ``` Reviewed By: fmassa Differential Revision: D26782757 fbshipit-source-id: 2039e1e6b4fea2b56bb4bcf2a017476f928e4928