[v1.x][LT] Add forward & backward linalg.gemm test for large size #18825

ChaiBapchya · 2020-07-30T01:08:29Z

Description

Tests mx.nd.linalg.gemm operator for large size tensor [not large dimension]

Results

 python3 -m pytest -s --exitfirst --verbose tests/nightly/test_large_array.py::test_linalg

============================================================================================================================= test session starts ==============================================================================================================================
platform linux -- Python 3.7.7, pytest-5.4.1, py-1.8.1, pluggy-0.13.1 -- /home/ubuntu/anaconda3/bin/python3
cachedir: .pytest_cache
hypothesis profile 'default' -> database=DirectoryBasedExampleDatabase('/home/ubuntu/chai-mxnet/.hypothesis/examples')
rootdir: /home/ubuntu/chai-mxnet
plugins: remotedata-0.3.2, openfiles-0.4.0, astropy-header-0.1.2, hypothesis-5.8.3, arraydiff-0.3, doctestplus-0.5.0
collected 1 item

tests/nightly/test_large_array.py::test_linalg [05:24:49] ../src/base.cc:84: Upgrade advisory: this mxnet has been built against cuDNN lib version 7501, which is older than the oldest version tested by CI (7600).  Set MXNET_CUDNN_LIB_CHECKING=0 to quiet this warning.
PASSED

=============================================================================================================================== warnings summary ===============================================================================================================================
/home/ubuntu/anaconda3/lib/python3.7/site-packages/nose/importer.py:12
  /home/ubuntu/anaconda3/lib/python3.7/site-packages/nose/importer.py:12: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses
    from imp import find_module, load_module, acquire_lock, release_lock

-- Docs: https://docs.pytest.org/en/latest/warnings.html
=================================================================================================================== 1 passed, 1 warning in 583.43s (0:09:43) ===================================================================================================================

Comments

FIxes indent in linalg.gemm2 test
Removes hardcode & replaces with "perturbation" for gradient checks

mxnet-bot · 2020-07-30T01:08:31Z

Hey @ChaiBapchya , Thanks for submitting the PR
All tests are already queued to run once. If tests fail, you can trigger one or more tests again with the following commands:

To trigger all jobs: @mxnet-bot run ci [all]
To trigger specific jobs: @mxnet-bot run ci [job1, job2]

CI supported jobs: [website, windows-cpu, sanity, centos-cpu, edge, miscellaneous, unix-cpu, clang, centos-gpu, windows-gpu, unix-gpu]

Note:
Only following 3 categories can trigger CI :PR Author, MXNet Committer, Jenkins Admin.
All CI tests must pass before the PR can be merged.

tests/nightly/test_large_array.py

access2rohit · 2020-07-30T05:16:15Z

left nitpicky comments. Rest LGTM !

access2rohit · 2020-07-30T06:38:33Z

tests/nightly/test_large_array.py

+                out = mx.nd.linalg.gemm(inp1, inp2, inp3, transpose_b=True)
+            return inp1.grad, inp2.grad, inp3.grad, out
+
+        inp1 = mx.nd.ones(shape=(MEDIUM_X, SMALL_Y, MEDIUM_X))


spaces around =

It's incorrect according to PEP8 style guide Python
https://www.python.org/dev/peps/pep-0008/

access2rohit · 2020-07-30T06:38:40Z

tests/nightly/test_large_array.py

+        inp1 = mx.nd.ones(shape=(MEDIUM_X, SMALL_Y, MEDIUM_X))
+        perturbation = 0.2
+        inp1[0][0][0] = perturbation
+        inp2 = mx.nd.ones(shape=(MEDIUM_X, SMALL_Y, MEDIUM_X))


spaces around =

It's incorrect according to PEP8 style guide Python
https://www.python.org/dev/peps/pep-0008/

access2rohit · 2020-07-30T06:38:47Z

tests/nightly/test_large_array.py

+        perturbation = 0.2
+        inp1[0][0][0] = perturbation
+        inp2 = mx.nd.ones(shape=(MEDIUM_X, SMALL_Y, MEDIUM_X))
+        inp3 = mx.nd.ones(shape=(MEDIUM_X, SMALL_Y, SMALL_Y))


It's incorrect according to PEP8 style guide Python
https://www.python.org/dev/peps/pep-0008/

ChaiBapchya · 2020-07-30T08:08:20Z

@mxnet-bot run ci [centos-cpu]

mxnet-bot · 2020-07-30T08:08:24Z

Jenkins CI successfully triggered : [centos-cpu]

add test for linalg.gemm

5b0ecae

access2rohit reviewed Jul 30, 2020

View reviewed changes

tests/nightly/test_large_array.py Outdated Show resolved Hide resolved

access2rohit reviewed Jul 30, 2020

View reviewed changes

tests/nightly/test_large_array.py Show resolved Hide resolved

access2rohit reviewed Jul 30, 2020

View reviewed changes

tests/nightly/test_large_array.py Outdated Show resolved Hide resolved

fix indents

e4f67fb

access2rohit reviewed Jul 30, 2020

View reviewed changes

ChaiBapchya mentioned this pull request Jul 30, 2020

flaky test: test_gluon_data_vision.test_random_transforms #17369

Open

Zha0q1 approved these changes Jul 30, 2020

View reviewed changes

access2rohit approved these changes Jul 30, 2020

View reviewed changes

sandeep-krishnamurthy merged commit 84c9e0d into apache:v1.x Jul 30, 2020

ChaiBapchya deleted the add_lt_gemm_test branch July 30, 2020 19:55

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[v1.x][LT] Add forward & backward linalg.gemm test for large size #18825

[v1.x][LT] Add forward & backward linalg.gemm test for large size #18825

ChaiBapchya commented Jul 30, 2020 •

edited

mxnet-bot commented Jul 30, 2020

access2rohit commented Jul 30, 2020

access2rohit Jul 30, 2020

ChaiBapchya Jul 30, 2020

access2rohit Jul 30, 2020

ChaiBapchya Jul 30, 2020

access2rohit Jul 30, 2020

ChaiBapchya Jul 30, 2020

ChaiBapchya commented Jul 30, 2020

mxnet-bot commented Jul 30, 2020

[v1.x][LT] Add forward & backward linalg.gemm test for large size #18825

[v1.x][LT] Add forward & backward linalg.gemm test for large size #18825

Conversation

ChaiBapchya commented Jul 30, 2020 • edited

Description

Results

Comments

mxnet-bot commented Jul 30, 2020

access2rohit commented Jul 30, 2020

access2rohit Jul 30, 2020

Choose a reason for hiding this comment

ChaiBapchya Jul 30, 2020

Choose a reason for hiding this comment

access2rohit Jul 30, 2020

Choose a reason for hiding this comment

ChaiBapchya Jul 30, 2020

Choose a reason for hiding this comment

access2rohit Jul 30, 2020

Choose a reason for hiding this comment

ChaiBapchya Jul 30, 2020

Choose a reason for hiding this comment

ChaiBapchya commented Jul 30, 2020

mxnet-bot commented Jul 30, 2020

ChaiBapchya commented Jul 30, 2020 •

edited