Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[tests] Fix bug in weights tensor layout in solver test #1950

Closed
wants to merge 6 commits into from

Conversation

xinlipn
Copy link
Contributor

@xinlipn xinlipn commented Jan 28, 2023

Weight tensor should be initilized as k, C, y, x instead of k, C, x, y

@junliume
Copy link
Contributor

@xinlipn please add an issue where this problem is found, and add a test :)

@junliume junliume requested a review from JehandadKhan January 28, 2023 07:40
@xinlipn
Copy link
Contributor Author

xinlipn commented Jan 30, 2023

@xinlipn please add an issue where this problem is found, and add a test :)
Hi @junliume , this was discovered while @JehandadKhan was reviewing another PR#22 on internal branch. This itself is a test :)

@xinlipn
Copy link
Contributor Author

xinlipn commented Feb 2, 2023

Local test seems to be ok. Will try to reproduce on different nodes.

130/136 Test #130: LOG_TEST.AssertLogCmdOutput_Neg ........................................................................................................... Passed 0.08 sec
Start 131: LOG_TEST.AssertLogFindCmdOutput_Neg
131/136 Test #131: LOG_TEST.AssertLogFindCmdOutput_Neg ....................................................................................................... Passed 0.08 sec
Start 132: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:16C:128H:16W:16k:128y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)
132/136 Test #132: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:16C:128H:16W:16k:128y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ......... Passed 0.25 sec
Start 133: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:128H:28W:28k:128y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)
133/136 Test #133: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:128H:28W:28k:128y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ......... Passed 0.49 sec
Start 134: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:256H:14W:14k:256y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)
134/136 Test #134: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:256H:14W:14k:256y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ......... Passed 0.38 sec
Start 135: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:512H:7W:7k:512y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)
135/136 Test #135: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:512H:7W:7k:512y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ........... Passed 0.41 sec
Start 136: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:1024H:14W:14k:1024y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)
136/136 Test #136: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:1024H:14W:14k:1024y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ....... Passed 1.15 sec

100% tests passed, 0 tests failed out of 85
Total Test time (real) = 127.38 sec

@xinlipn
Copy link
Contributor Author

xinlipn commented Feb 3, 2023

Tested on gfx908 node and also passed.

    Start 133: LOG_TEST.AssertLogFindCmdOutput_Neg

133/138 Test #133: LOG_TEST.AssertLogFindCmdOutput_Neg ....................................................................................................... Passed 0.10 sec
Start 134: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:16C:128H:16W:16k:128y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)
134/138 Test #134: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:16C:128H:16W:16k:128y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ......... Passed 1.88 sec
Start 135: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:128H:28W:28k:128y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)
135/138 Test #135: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:128H:28W:28k:128y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ......... Passed 13.49 sec
Start 136: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:256H:14W:14k:256y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)
136/138 Test #136: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:256H:14W:14k:256y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ......... Passed 12.31 sec
Start 137: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:512H:7W:7k:512y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)
137/138 Test #137: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:512H:7W:7k:512y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ........... Passed 11.86 sec
Start 138: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:1024H:14W:14k:1024y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0)

138/138 Test #138: ConvFwdTest/ConvFwdSolverTestFloat.ConvASM3x3UFwd/(1,N:64C:1024H:14W:14k:1024y:3x:3pad_y:1pad_x:1stride_y:1dilation_y:1conv_mode:0) ....... Passed 178.18 sec

100% tests passed, 0 tests failed out of 87

Total Test time (real) = 393.76 sec

@atamazov
Copy link
Contributor

atamazov commented Feb 5, 2023

@xinlipn Please format logs like this:

```
log text (will be rendered in monospaced font, no word wrapping)
```

When logs are very long, either attach as text files or consult https://github.com/ROCmSoftwarePlatform/MIOpen/wiki/How-to-insert-console-logs-into-github-pages

@atamazov
Copy link
Contributor

atamazov commented Feb 5, 2023

@xinlipn Please add [tests] to the title of this PR, thanks.

@xinlipn xinlipn changed the title Fix bug in weights tensor layout in solver test [tests] Fix bug in weights tensor layout in solver test Feb 6, 2023
@xinlipn xinlipn marked this pull request as draft February 7, 2023 21:48
@@ -99,7 +99,7 @@ struct ConvFwdSolverTest
test_skipped = false;
std::tie(algo, conv_config) = GetParam();
input = tensor<T>{conv_config.N, conv_config.C, conv_config.H, conv_config.W};
weights = tensor<T>{1, conv_config.k, conv_config.x, conv_config.y};
weights = tensor<T>{conv_config.k, conv_config.C, conv_config.x, conv_config.y};
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
weights = tensor<T>{conv_config.k, conv_config.C, conv_config.x, conv_config.y};
weights = tensor<T>{conv_config.k, conv_config.C, conv_config.y, conv_config.x};

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@xinlipn xinlipn marked this pull request as ready for review September 11, 2023 07:05
@xinlipn
Copy link
Contributor Author

xinlipn commented Oct 25, 2023

The original file was renamed. A new PR had been created, reviewed and merged. Thus close this PR

#2436

@xinlipn xinlipn closed this Oct 25, 2023
@junliume junliume deleted the sl/solver_test_bug branch June 25, 2024 05:05
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants