Move THCTensor_(cauchy) to ATen #21289
Closed
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This reverts commit dc8f306.
pytorchbot
added
module: cuda
Contributor
facebook-github-bot
left a comment
facebook-github-bot
left a comment
There was a problem hiding this comment.
@ezyang is landing this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
ezyang
approved these changes
Jun 4, 2019
zdevito
pushed a commit
to zdevito/ATen
that referenced
this pull request
Jun 4, 2019
Summary: ## Effective Bandwidth Benchmark - using https://gist.github.com/syed-ahmed/f8b7384d642f4bce484228b508b4bc68 - on V100 ### Float Type #### Before: ``` cauchy, size, elements 65536 forward 4.980564117431641e-06 bandwidth (GB/s) 52.63339529803734 cauchy, size, elements 131072 forward 6.232261657714844e-06 bandwidth (GB/s) 84.12483762631982 cauchy, size, elements 262144 forward 9.548664093017577e-06 bandwidth (GB/s) 109.81389540833959 cauchy, size, elements 524288 forward 1.59454345703125e-05 bandwidth (GB/s) 131.52052963827754 cauchy, size, elements 1048576 forward 2.86865234375e-05 bandwidth (GB/s) 146.21165262978724 cauchy, size, elements 2097152 forward 5.4748058319091796e-05 bandwidth (GB/s) 153.2220184158516 cauchy, size, elements 4194304 forward 0.00010075807571411133 bandwidth (GB/s) 166.50988897012377 cauchy, size, elements 8388608 forward 0.0001935744285583496 bandwidth (GB/s) 173.34124269355965 cauchy, size, elements 16777216 forward 0.00038077831268310545 bandwidth (GB/s) 176.24129779641603 cauchy, size, elements 33554432 forward 0.0006851387023925781 bandwidth (GB/s) 195.8986224705994 ``` #### After: ``` cauchy, size, elements 65536 forward 6.077289581298828e-06 bandwidth (GB/s) 43.13501874366419 cauchy, size, elements 131072 forward 6.2131881713867184e-06 bandwidth (GB/s) 84.38308731972373 cauchy, size, elements 262144 forward 6.46829605102539e-06 bandwidth (GB/s) 162.11008150033175 cauchy, size, elements 524288 forward 6.8783760070800785e-06 bandwidth (GB/s) 304.8905726935182 cauchy, size, elements 1048576 forward 9.505748748779296e-06 bandwidth (GB/s) 441.23867681003264 cauchy, size, elements 2097152 forward 1.5070438385009766e-05 bandwidth (GB/s) 556.6266744001266 cauchy, size, elements 4194304 forward 2.4406909942626954e-05 bandwidth (GB/s) 687.396152951685 cauchy, size, elements 8388608 forward 4.6243667602539064e-05 bandwidth (GB/s) 725.6005792706125 cauchy, size, elements 16777216 forward 9.100198745727539e-05 bandwidth (GB/s) 737.4439380404413 cauchy, size, elements 33554432 forward 0.00017449140548706055 bandwidth (GB/s) 769.1939188944922 ``` ### Double Type #### Before: ``` cauchy, size, elements 65536 forward 4.885196685791015e-06 bandwidth (GB/s) 53.660889593753055 cauchy, size, elements 131072 forward 6.229877471923828e-06 bandwidth (GB/s) 84.15703235943361 cauchy, size, elements 262144 forward 9.605884552001953e-06 bandwidth (GB/s) 109.15975455706132 cauchy, size, elements 524288 forward 1.5976428985595704e-05 bandwidth (GB/s) 131.26537863315923 cauchy, size, elements 1048576 forward 2.9621124267578124e-05 bandwidth (GB/s) 141.59840666786866 cauchy, size, elements 2097152 forward 5.5103302001953126e-05 bandwidth (GB/s) 152.23421637604707 cauchy, size, elements 4194304 forward 0.00010124444961547851 bandwidth (GB/s) 165.70998275677383 cauchy, size, elements 8388608 forward 0.0001944279670715332 bandwidth (GB/s) 172.58027487195184 cauchy, size, elements 16777216 forward 0.00034950494766235353 bandwidth (GB/s) 192.01119883668116 cauchy, size, elements 33554432 forward 0.0007002186775207519 bandwidth (GB/s) 191.67973135938277 ``` #### After: ``` cauchy, size, elements 65536 forward 5.91278076171875e-06 bandwidth (GB/s) 44.33514628129032 cauchy, size, elements 131072 forward 6.234645843505859e-06 bandwidth (GB/s) 84.09266751632889 cauchy, size, elements 262144 forward 7.433891296386719e-06 bandwidth (GB/s) 141.05344807902503 cauchy, size, elements 524288 forward 1.1401176452636719e-05 bandwidth (GB/s) 183.94171941045587 cauchy, size, elements 1048576 forward 1.960039138793945e-05 bandwidth (GB/s) 213.99082890665372 cauchy, size, elements 2097152 forward 3.434181213378906e-05 bandwidth (GB/s) 244.26806504326578 cauchy, size, elements 4194304 forward 6.517410278320313e-05 bandwidth (GB/s) 257.4215107465028 cauchy, size, elements 8388608 forward 0.0001229524612426758 bandwidth (GB/s) 272.9057365819818 cauchy, size, elements 16777216 forward 0.00023239374160766602 bandwidth (GB/s) 288.77225150621814 cauchy, size, elements 33554432 forward 0.00046050310134887696 bandwidth (GB/s) 291.4589013773367 ``` Resubmit of pytorch/pytorch#20622 Pull Request resolved: pytorch/pytorch#21289 Differential Revision: D15622713 Pulled By: ezyang fbshipit-source-id: abe8bd57794bd1c3a0b92395367a9653c5d0f2db
Chillee
added a commit
that referenced
this pull request
Jun 4, 2019
Summary: ## Effective Bandwidth Benchmark - using https://gist.github.com/syed-ahmed/f8b7384d642f4bce484228b508b4bc68 - on V100 ### Float Type #### Before: ``` cauchy, size, elements 65536 forward 4.980564117431641e-06 bandwidth (GB/s) 52.63339529803734 cauchy, size, elements 131072 forward 6.232261657714844e-06 bandwidth (GB/s) 84.12483762631982 cauchy, size, elements 262144 forward 9.548664093017577e-06 bandwidth (GB/s) 109.81389540833959 cauchy, size, elements 524288 forward 1.59454345703125e-05 bandwidth (GB/s) 131.52052963827754 cauchy, size, elements 1048576 forward 2.86865234375e-05 bandwidth (GB/s) 146.21165262978724 cauchy, size, elements 2097152 forward 5.4748058319091796e-05 bandwidth (GB/s) 153.2220184158516 cauchy, size, elements 4194304 forward 0.00010075807571411133 bandwidth (GB/s) 166.50988897012377 cauchy, size, elements 8388608 forward 0.0001935744285583496 bandwidth (GB/s) 173.34124269355965 cauchy, size, elements 16777216 forward 0.00038077831268310545 bandwidth (GB/s) 176.24129779641603 cauchy, size, elements 33554432 forward 0.0006851387023925781 bandwidth (GB/s) 195.8986224705994 ``` #### After: ``` cauchy, size, elements 65536 forward 6.077289581298828e-06 bandwidth (GB/s) 43.13501874366419 cauchy, size, elements 131072 forward 6.2131881713867184e-06 bandwidth (GB/s) 84.38308731972373 cauchy, size, elements 262144 forward 6.46829605102539e-06 bandwidth (GB/s) 162.11008150033175 cauchy, size, elements 524288 forward 6.8783760070800785e-06 bandwidth (GB/s) 304.8905726935182 cauchy, size, elements 1048576 forward 9.505748748779296e-06 bandwidth (GB/s) 441.23867681003264 cauchy, size, elements 2097152 forward 1.5070438385009766e-05 bandwidth (GB/s) 556.6266744001266 cauchy, size, elements 4194304 forward 2.4406909942626954e-05 bandwidth (GB/s) 687.396152951685 cauchy, size, elements 8388608 forward 4.6243667602539064e-05 bandwidth (GB/s) 725.6005792706125 cauchy, size, elements 16777216 forward 9.100198745727539e-05 bandwidth (GB/s) 737.4439380404413 cauchy, size, elements 33554432 forward 0.00017449140548706055 bandwidth (GB/s) 769.1939188944922 ``` ### Double Type #### Before: ``` cauchy, size, elements 65536 forward 4.885196685791015e-06 bandwidth (GB/s) 53.660889593753055 cauchy, size, elements 131072 forward 6.229877471923828e-06 bandwidth (GB/s) 84.15703235943361 cauchy, size, elements 262144 forward 9.605884552001953e-06 bandwidth (GB/s) 109.15975455706132 cauchy, size, elements 524288 forward 1.5976428985595704e-05 bandwidth (GB/s) 131.26537863315923 cauchy, size, elements 1048576 forward 2.9621124267578124e-05 bandwidth (GB/s) 141.59840666786866 cauchy, size, elements 2097152 forward 5.5103302001953126e-05 bandwidth (GB/s) 152.23421637604707 cauchy, size, elements 4194304 forward 0.00010124444961547851 bandwidth (GB/s) 165.70998275677383 cauchy, size, elements 8388608 forward 0.0001944279670715332 bandwidth (GB/s) 172.58027487195184 cauchy, size, elements 16777216 forward 0.00034950494766235353 bandwidth (GB/s) 192.01119883668116 cauchy, size, elements 33554432 forward 0.0007002186775207519 bandwidth (GB/s) 191.67973135938277 ``` #### After: ``` cauchy, size, elements 65536 forward 5.91278076171875e-06 bandwidth (GB/s) 44.33514628129032 cauchy, size, elements 131072 forward 6.234645843505859e-06 bandwidth (GB/s) 84.09266751632889 cauchy, size, elements 262144 forward 7.433891296386719e-06 bandwidth (GB/s) 141.05344807902503 cauchy, size, elements 524288 forward 1.1401176452636719e-05 bandwidth (GB/s) 183.94171941045587 cauchy, size, elements 1048576 forward 1.960039138793945e-05 bandwidth (GB/s) 213.99082890665372 cauchy, size, elements 2097152 forward 3.434181213378906e-05 bandwidth (GB/s) 244.26806504326578 cauchy, size, elements 4194304 forward 6.517410278320313e-05 bandwidth (GB/s) 257.4215107465028 cauchy, size, elements 8388608 forward 0.0001229524612426758 bandwidth (GB/s) 272.9057365819818 cauchy, size, elements 16777216 forward 0.00023239374160766602 bandwidth (GB/s) 288.77225150621814 cauchy, size, elements 33554432 forward 0.00046050310134887696 bandwidth (GB/s) 291.4589013773367 ``` Resubmit of #20622 Pull Request resolved: #21289 Differential Revision: D15622713 Pulled By: ezyang fbshipit-source-id: abe8bd57794bd1c3a0b92395367a9653c5d0f2db
Contributor
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Effective Bandwidth Benchmark
Float Type
Before:
After:
Double Type
Before:
After:
Resubmit of #20622