lebeg commented on issue #12203: flaky test: test_operator_gpu.test_depthwise_convolution URL: https://github.com/apache/incubator-mxnet/issues/12203#issuecomment-449337334 @juliusshufan We use dockerized builds and tests and therefore the host system shouldn't matter. You should be able to reproduce the failure by following steps mentioned by @mseth10 above: **Checkout and build** ```bash git clone --recursive https://github.com/apache/incubator-mxnet.git cd incubator-mxnet pip3 install -r ci/requirements.txt ci/build.py --platform ubuntu_build_cuda /work/runtime_functions.sh build_ubuntu_gpu_mkldnn ``` **Enable the test** Comment out [this line](https://github.com/apache/incubator-mxnet/blob/master/tests/python/unittest/test_operator.py#L1634) in file [tests/python/unittest/test_operator.py](https://github.com/apache/incubator-mxnet/blob/master/tests/python/unittest/test_operator.py): ```python # @unittest.skip("Flaky test https://github.com/apache/incubator-mxnet/issues/12203";) ``` **Speed up the testing** Running only this particular test 10,000 times: Modify the [unittest_ubuntu_python2_gpu](https://github.com/apache/incubator-mxnet/blob/master/ci/docker/runtime_functions.sh#L779-L786) function in file [ci/docker/runtime_functions.sh](https://github.com/apache/incubator-mxnet/blob/master/ci/docker/runtime_functions.sh): ```bash MXNET_TEST_COUNT=10000 nosetests-2.7 $NOSE_COVERAGE_ARGUMENTS $NOSE_TIMER_ARGUMENTS --with-xunit --xunit-file nosetests_gpu.xml --verbose tests/python/gpu/test_operator_gpu.py:test_depthwise_convolution ``` **Run the test** In the exact environment where it fails: ``` ci/build.py --nvidiadocker --platform ubuntu_gpu /work/runtime_functions.sh unittest_ubuntu_python2_gpu ```
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
