zhaoyang-star opened a new issue, #13154: URL: https://github.com/apache/tvm/issues/13154
The unittest `test_pipeline_executor.py` runs right. But when I change [Line 473](https://github.com/apache/tvm/blob/main/tests/python/relay/test_pipeline_executor.py#L473) to `datas.append(np.random.randn(*dshape))`, the unittest will fail as following: ``` np.testing.assert_allclose(actual.shape, desired.shape) > np.testing.assert_allclose(actual, desired, rtol=rtol, atol=atol, verbose=True) E AssertionError: E Not equal to tolerance rtol=1e-07, atol=1e-07 E E Mismatched elements: 9 / 9 (100%) E Max absolute difference: 1.34077008 E Max relative difference: 2.36375847e+44 E x: array([[-1.34077 , 1.324932, -0.255464], E [-0.332742, 0.648831, -0.631121], E [ 0.561325, -1.151209, -0.033365]]) E y: array([[0.e+00, 6.e-45, 0.e+00], E [0.e+00, 3.e-45, 0.e+00], E [0.e+00, 0.e+00, 0.e+00]], dtype=float32) python/tvm/testing/utils.py:114: AssertionError ``` So I want to confirm that Multi-batch has not be supported in Pipeline Executor yet？Thanks for your advice @huajsj @comaniac -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
