leezu commented on issue #12287: Add stable nrm2 for L2_normalization URL: https://github.com/apache/incubator-mxnet/pull/12287#issuecomment-416237810 @TccccD there seems to be a bug in the current version of this PR leading to `Shape size mismatch 3 v.s. 15`: ``` ====================================================================== ERROR: test_operator.test_l2_normalization ---------------------------------------------------------------------- Traceback (most recent call last): File "C:\Anaconda3\envs\py2\lib\site-packages\nose\case.py", line 197, in runTest self.test(*self.arg) File "C:\jenkins_slave\workspace\ut-python-gpu\tests\python\unittest\common.py", line 172, in test_new orig_test(*args, **kwargs) File "C:\jenkins_slave\workspace\ut-python-gpu\tests\python\unittest\test_operator.py", line 3135, in test_l2_normalization check_l2_normalization((nbatch, nchannel, height), mode, dtype) File "C:\jenkins_slave\workspace\ut-python-gpu\tests\python\unittest\test_operator.py", line 3123, in check_l2_normalization check_numeric_gradient(out, [in_data], numeric_eps=1e-3, rtol=1e-2, atol=1e-3) File "C:\jenkins_slave\workspace\ut-python-gpu\windows_package\python\mxnet\test_utils.py", line 900, in check_numeric_gradient symbolic_grads = {k:executor.grad_dict[k].asnumpy() for k in grad_nodes} File "C:\jenkins_slave\workspace\ut-python-gpu\windows_package\python\mxnet\test_utils.py", line 900, in <dictcomp> symbolic_grads = {k:executor.grad_dict[k].asnumpy() for k in grad_nodes} File "C:\jenkins_slave\workspace\ut-python-gpu\windows_package\python\mxnet\ndarray\ndarray.py", line 1972, in asnumpy ctypes.c_size_t(data.size))) File "C:\jenkins_slave\workspace\ut-python-gpu\windows_package\python\mxnet\base.py", line 252, in check_call raise MXNetError(py_str(_LIB.MXGetLastError())) MXNetError: [12:50:19] c:\jenkins_slave\workspace\build-gpu@3\include\mxnet\./tensor_blob.h:146: Check failed: this->shape_.Size() == shape.Size() (3 vs. 15) Shape size mismatch 3 v.s. 15 ```
---------------------------------------------------------------- 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
