{ai}[foss/2022a] PyTorch v1.13.1 w/ Python 3.10.4 #17155

branfosj · 2023-01-19T09:31:48Z

(created using eb --new-pr)

Requires

{tools}[GCCcore/11.3.0] pytest-rerunfailures v11.1, pytest-shard v0.1.2 w/ Python 3.10.4 #17295

…-1.13.1_fix-test-ops-conf.patch, PyTorch-1.13.1_no-cuda-stubs-rpath.patch, PyTorch-1.13.1_remove-flaky-test-in-testnn.patch, PyTorch-1.13.1_skip-ao-sparsity-test-without-fbgemm.patch

Update patches based on PyTorch 1.13.1

Those tests require 2 pytest plugins and a bugfix.

Fix test_ops* startup failures

…asyconfigs into 20230119093142_new_pr_PyTorch1131

branfosj · 2023-02-10T17:26:18Z

@boegelbot please test @ jsc-zen2

boegelbot · 2023-02-10T17:27:11Z

@branfosj: Request for testing this PR well received on jsczen2l1.int.jsc-zen2.easybuild-test.cluster

PR test command 'EB_PR=17155 EB_ARGS= /opt/software/slurm/bin/sbatch --mem-per-cpu=4000M --job-name test_PR_17155 --ntasks=8 ~/boegelbot/eb_from_pr_upload_jsc-zen2.sh' executed!

exit code: 0
output:

Submitted batch job 2179

Test results coming soon (I hope)...

- notification for comment with ID 1426118567 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

branfosj · 2023-02-10T18:59:58Z

Test report by @branfosj
FAILED
Build succeeded for 0 out of 1 (1 easyconfigs in total)
bear-pg0105u03a.bear.cluster - Linux RHEL 8.6, x86_64, Intel(R) Xeon(R) Platinum 8360Y CPU @ 2.40GHz (icelake), Python 3.6.8
See https://gist.github.com/38b9646d67afe7ea346ceaa169839769 for a full test report.

branfosj · 2023-02-10T19:05:51Z

TestGradientsCPU.test_forward_mode_AD_nn_functional_max_unpool2d_cpu_float64
Unexpected success

I've lost PyTorch-1.12.1_skip-failing-grad-test.patch from the patch list.

Flamefire · 2023-02-10T19:15:08Z

Test report by @Flamefire
FAILED
Build succeeded for 0 out of 1 (1 easyconfigs in total)
taurusi8009 - Linux CentOS Linux 7.9.2009, x86_64, AMD EPYC 7352 24-Core Processor (zen2), 8 x NVIDIA NVIDIA A100-SXM4-40GB, 470.57.02, Python 2.7.5
See https://gist.github.com/0f4c49467e073b6697de4a6dd50923b2 for a full test report.

branfosj · 2023-02-10T20:39:20Z

Test report by @branfosj
FAILED
Build succeeded for 8 out of 9 (1 easyconfigs in total)
bear-pg0210u06b.bear.cluster - Linux RHEL 8.6, x86_64, Intel(R) Xeon(R) CPU E5-2640 v4 @ 2.40GHz (broadwell), Python 3.6.8
See https://gist.github.com/bc0482f9ab1a4f51e2ed820ecf3deaf6 for a full test report.

Flamefire · 2023-02-10T21:49:48Z

Test report by @Flamefire
FAILED
Build succeeded for 0 out of 1 (1 easyconfigs in total)
taurusml24 - Linux RHEL 7.6, POWER, 8335-GTX (power9le), 6 x NVIDIA Tesla V100-SXM2-32GB, 440.64.00, Python 2.7.5
See https://gist.github.com/564b9bf8970d481af146ddcb73b16cec for a full test report.

Flamefire · 2023-02-10T22:48:37Z

Test report by @Flamefire
FAILED
Build succeeded for 0 out of 1 (1 easyconfigs in total)
taurusa12 - Linux CentOS Linux 7.7.1908, x86_64, Intel(R) Xeon(R) CPU E5-2603 v4 @ 1.70GHz (broadwell), 3 x NVIDIA GeForce GTX 1080 Ti, 460.32.03, Python 2.7.5
See https://gist.github.com/6247717cb51697c6752ae381309b3d79 for a full test report.

boegelbot · 2023-02-11T00:51:47Z

Test report by @boegelbot
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
jsczen2c1.int.jsc-zen2.easybuild-test.cluster - Linux Rocky Linux 8.5, x86_64, AMD EPYC 7742 64-Core Processor (zen2), Python 3.6.8
See https://gist.github.com/3df87c14f170259201e78bb1620a10f1 for a full test report.

branfosj · 2023-02-11T09:55:44Z

Patches added:

test_ops_gradients: I'd missed adding PyTorch-1.13.1_skip-failing-grad-test.patch and that updated patch covers the unexpected successes that I see
test_cpp_extensions_aot_ninja (and test_cpp_extensions_aot_no_ninja?): PyTorch-1.13.1_install-vsx-vec-headers.patch should fix these
test_ops: PyTorch-1.13.1_increase-tolerance-test_ops.patch

To investigate:

test_ao_sparsity - POWER only
test_quantization - POWER only
distributed/rpc/test_tensorpipe_agent - POWER only
test_cpp_extensions_open_device_registration - POWER only
test_native_mha - I'm seeing this on broadwell only and also in PyTorch 1.12.1

branfosj · 2023-02-11T12:23:24Z

Test report by @branfosj
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
bear-pg0105u03a.bear.cluster - Linux RHEL 8.6, x86_64, Intel(R) Xeon(R) Platinum 8360Y CPU @ 2.40GHz (icelake), Python 3.6.8
See https://gist.github.com/edc885781ddaa241515bca898ae0bfab for a full test report.

boegel · 2023-02-16T09:17:42Z

@boegelbot please test @ generoso
CORE_CNT=16

boegelbot · 2023-02-16T09:20:09Z

@boegel: Request for testing this PR well received on login1

PR test command 'EB_PR=17155 EB_ARGS= EB_CONTAINER= /opt/software/slurm/bin/sbatch --job-name test_PR_17155 --ntasks="16" ~/boegelbot/eb_from_pr_upload_generoso.sh' executed!

exit code: 0
output:

Submitted batch job 10252

Test results coming soon (I hope)...

- notification for comment with ID 1432765219 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

branfosj · 2023-04-14T08:09:30Z

@branfosj distributed/rpc/test_tensorpipe_agent is failing, just like in #17156, so let's skip that too, and mention it in #17712?

Done in 52cbf0e

The generoso build also failed on distributed/rpc/test_tensorpipe_agent and hit the 502 error on uploading the test report.

branfosj · 2023-04-14T08:38:59Z

@boegelbot please test @ generoso
CORE_CNT=16

boegelbot · 2023-04-14T08:40:09Z

@branfosj: Request for testing this PR well received on login1

PR test command 'EB_PR=17155 EB_ARGS= EB_CONTAINER= /opt/software/slurm/bin/sbatch --job-name test_PR_17155 --ntasks="16" ~/boegelbot/eb_from_pr_upload_generoso.sh' executed!

exit code: 0
output:

Submitted batch job 10651

Test results coming soon (I hope)...

- notification for comment with ID 1508148644 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

branfosj · 2023-04-14T09:36:57Z

Test report by @branfosj
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
bear-pg0104u04a.bear.cluster - Linux RHEL 8.6, x86_64, Intel(R) Xeon(R) Platinum 8360Y CPU @ 2.40GHz (icelake), Python 3.6.8
See https://gist.github.com/branfosj/574855412ab5b55e20b9f3602df05ab9 for a full test report.

boegel · 2023-04-14T10:52:49Z

Test report by @boegel
FAILED
Build succeeded for 0 out of 1 (1 easyconfigs in total)
node3101.skitty.os - Linux RHEL 8.6, x86_64, Intel(R) Xeon(R) Gold 6140 CPU @ 2.30GHz (skylake_avx512), Python 3.6.8
See https://gist.github.com/boegel/51d67d8eff4c677e7dd06cd7a653d833 for a full test report.

VRehnberg · 2023-04-21T10:22:04Z

Test report by @VRehnberg
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
alvis-c1 - Linux Rocky Linux 8.6, x86_64, Intel Xeon Processor (Skylake), Python 3.6.8
See https://gist.github.com/VRehnberg/8a400e560a8b60563c219156171c18c6 for a full test report.

VRehnberg · 2023-04-25T12:59:38Z

Test report by @VRehnberg
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
alvis-s1 - Linux Rocky Linux 8.6, x86_64, Intel(R) Xeon(R) Silver 4314 CPU @ 2.40GHz, Python 3.6.8
See https://gist.github.com/VRehnberg/75c0c0548375006d6bb895020529a6d0 for a full test report.

boegel · 2023-07-05T17:10:04Z

Two tests were failing for me:

test_quantization failed!
distributed/test_c10d_gloo failed!

test_quantization should be fixed by patch PyTorch-1.12.1_add-hypothesis-suppression.patch) added in #17908
distributed/test_c10d_gloo should be fixed by patch PyTorch-1.12.1_skip-test_round_robin.patch added in #16793

boegel · 2023-07-05T17:17:59Z

@boegelbot please test @ generoso
CORE_CNT=16

boegelbot · 2023-07-05T17:20:09Z

@boegel: Request for testing this PR well received on login1

PR test command 'EB_PR=17155 EB_ARGS= EB_CONTAINER= /opt/software/slurm/bin/sbatch --job-name test_PR_17155 --ntasks="16" ~/boegelbot/eb_from_pr_upload_generoso.sh' executed!

exit code: 0
output:

Submitted batch job 11214

Test results coming soon (I hope)...

- notification for comment with ID 1622173121 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

boegel · 2023-07-05T19:13:11Z

@boegelbot please test @ jsc-zen2
CORE_CNT=16

boegelbot · 2023-07-05T19:17:08Z

@boegel: Request for testing this PR well received on jsczen2l1.int.jsc-zen2.easybuild-test.cluster

PR test command 'EB_PR=17155 EB_ARGS= /opt/software/slurm/bin/sbatch --mem-per-cpu=4000M --job-name test_PR_17155 --ntasks="16" ~/boegelbot/eb_from_pr_upload_jsc-zen2.sh' executed!

exit code: 0
output:

Submitted batch job 2956

Test results coming soon (I hope)...

- notification for comment with ID 1622324495 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

boegelbot · 2023-07-05T20:00:47Z

Test report by @boegelbot
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
cnx1 - Linux Rocky Linux 8.5, x86_64, Intel(R) Xeon(R) CPU E5-2690 v3 @ 2.60GHz (haswell), Python 3.6.8
See https://gist.github.com/boegelbot/ed9e61ab9cc559265403338028d3769e for a full test report.

branfosj · 2023-07-05T21:01:49Z

Test report by @branfosj
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
bear-pg0105u03a.bear.cluster - Linux RHEL 8.6, x86_64, Intel(R) Xeon(R) Platinum 8360Y CPU @ 2.40GHz (icelake), Python 3.6.8
See https://gist.github.com/branfosj/49deeeaa7871b8f65f61fe65b89d9cd1 for a full test report.

boegelbot · 2023-07-05T21:30:11Z

Test report by @boegelbot
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
jsczen2c1.int.jsc-zen2.easybuild-test.cluster - Linux Rocky Linux 8.5, x86_64, AMD EPYC 7742 64-Core Processor (zen2), Python 3.6.8
See https://gist.github.com/boegelbot/1a5b106a4ce60edddefc08041dd851f4 for a full test report.

boegel · 2023-07-06T04:40:00Z

Test report by @boegel
SUCCESS
Build succeeded for 1 out of 1 (1 easyconfigs in total)
node3108.skitty.os - Linux RHEL 8.6, x86_64, Intel(R) Xeon(R) Gold 6140 CPU @ 2.30GHz (skylake_avx512), Python 3.6.8
See https://gist.github.com/boegel/a1fd23ab14e246966ae037d4981a1285 for a full test report.

boegel · 2023-07-06T07:32:08Z

Going in, thanks @branfosj!

adding easyconfigs: PyTorch-1.13.1-foss-2022a.eb and patches: PyTorch…

6fac76c

…-1.13.1_fix-test-ops-conf.patch, PyTorch-1.13.1_no-cuda-stubs-rpath.patch, PyTorch-1.13.1_remove-flaky-test-in-testnn.patch, PyTorch-1.13.1_skip-ao-sparsity-test-without-fbgemm.patch

branfosj added the update label Jan 19, 2023

branfosj mentioned this pull request Jan 19, 2023

{ai}[foss/2022a] PyTorch v1.13.1 w/ Python 3.10.4 w/ CUDA 11.7.0 #17156

Closed

2 tasks

branfosj marked this pull request as draft January 19, 2023 10:09

This comment was marked as outdated.

Sign in to view

Flamefire and others added 3 commits February 10, 2023 11:43

Update patches based on PyTorch 1.13.1

3f6e57d

Merge pull request #4 from Flamefire/PyTorch-1.13.1

915994d

Update patches based on PyTorch 1.13.1

Fix test_ops* startup failures

5f52236

Those tests require 2 pytest plugins and a bugfix.

This was referenced Feb 10, 2023

Fix test_ops* startup failures branfosj/easybuild-easyconfigs#6

Merged

Fix test_ops* startup failures branfosj/easybuild-easyconfigs#7

Merged

Merge pull request #7 from Flamefire/PyTorch-1.13.1

041d5a5

Fix test_ops* startup failures

This comment was marked as outdated.

Sign in to view

Merge branch 'develop' of https://github.com/easybuilders/easybuild-e…

fc45081

…asyconfigs into 20230119093142_new_pr_PyTorch1131

branfosj marked this pull request as ready for review February 10, 2023 16:54

more patches

6124d4c

boegel added this to the 4.x milestone Feb 16, 2023

skip distributed/rpc/test_tensorpipe_agent

52cbf0e

boegel modified the milestones: next release (4.7.2), release after 4.7.2 May 23, 2023

boegel modified the milestones: 4.7.3, release after 4.7.3 Jul 5, 2023

boegel added 2 commits July 5, 2023 19:12

Merge branch 'develop' into 20230119093142_new_pr_PyTorch1131

c79ebe4

add extra patch files for PyTorch 1.13.1 to fix failing tests

233902d

boegel approved these changes Jul 6, 2023

View reviewed changes

boegel merged commit cdc3499 into easybuilders:develop Jul 6, 2023

boegel modified the milestones: release after 4.7.3, 4.7.3 Jul 6, 2023

branfosj deleted the 20230119093142_new_pr_PyTorch1131 branch July 6, 2023 07:36

Flamefire mentioned this pull request Aug 2, 2023

add/fix patches for PyTorch 1.13.1 w/ foss/2022a #18371

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

{ai}[foss/2022a] PyTorch v1.13.1 w/ Python 3.10.4 #17155

{ai}[foss/2022a] PyTorch v1.13.1 w/ Python 3.10.4 #17155

branfosj commented Jan 19, 2023 •

edited

Loading

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

branfosj commented Feb 10, 2023

boegelbot commented Feb 10, 2023

branfosj commented Feb 10, 2023

branfosj commented Feb 10, 2023 •

edited

Loading

Flamefire commented Feb 10, 2023

branfosj commented Feb 10, 2023

Flamefire commented Feb 10, 2023

Flamefire commented Feb 10, 2023

boegelbot commented Feb 11, 2023

branfosj commented Feb 11, 2023 •

edited

Loading

branfosj commented Feb 11, 2023

boegel commented Feb 16, 2023

boegelbot commented Feb 16, 2023

branfosj commented Apr 14, 2023

branfosj commented Apr 14, 2023

boegelbot commented Apr 14, 2023

branfosj commented Apr 14, 2023

boegel commented Apr 14, 2023

VRehnberg commented Apr 21, 2023

VRehnberg commented Apr 25, 2023

boegel commented Jul 5, 2023

boegel commented Jul 5, 2023

boegelbot commented Jul 5, 2023

boegel commented Jul 5, 2023

boegelbot commented Jul 5, 2023

boegelbot commented Jul 5, 2023

branfosj commented Jul 5, 2023

boegelbot commented Jul 5, 2023

boegel commented Jul 6, 2023

boegel commented Jul 6, 2023

{ai}[foss/2022a] PyTorch v1.13.1 w/ Python 3.10.4 #17155

{ai}[foss/2022a] PyTorch v1.13.1 w/ Python 3.10.4 #17155

Conversation

branfosj commented Jan 19, 2023 • edited Loading

This comment was marked as outdated.

This comment was marked as outdated.

This comment was marked as outdated.

branfosj commented Feb 10, 2023

boegelbot commented Feb 10, 2023

branfosj commented Feb 10, 2023

branfosj commented Feb 10, 2023 • edited Loading

Flamefire commented Feb 10, 2023

branfosj commented Feb 10, 2023

Flamefire commented Feb 10, 2023

Flamefire commented Feb 10, 2023

boegelbot commented Feb 11, 2023

branfosj commented Feb 11, 2023 • edited Loading

branfosj commented Feb 11, 2023

boegel commented Feb 16, 2023

boegelbot commented Feb 16, 2023

branfosj commented Apr 14, 2023

branfosj commented Apr 14, 2023

boegelbot commented Apr 14, 2023

branfosj commented Apr 14, 2023

boegel commented Apr 14, 2023

VRehnberg commented Apr 21, 2023

VRehnberg commented Apr 25, 2023

boegel commented Jul 5, 2023

boegel commented Jul 5, 2023

boegelbot commented Jul 5, 2023

boegel commented Jul 5, 2023

boegelbot commented Jul 5, 2023

boegelbot commented Jul 5, 2023

branfosj commented Jul 5, 2023

boegelbot commented Jul 5, 2023

boegel commented Jul 6, 2023

boegel commented Jul 6, 2023

branfosj commented Jan 19, 2023 •

edited

Loading

branfosj commented Feb 10, 2023 •

edited

Loading

branfosj commented Feb 11, 2023 •

edited

Loading