lionel92 opened a new issue #10807: Asnumpy() error with gluon dense under both GPU/CPU/MKLDNN environment URL: https://github.com/apache/incubator-mxnet/issues/10807 ## Description I try to use gluon API to test mkldnn while an error occurs when do asnumpy() operation after the network's backward propogation. So I performed the test under gpu and native cpu environment, and the error still exists. You can run the following mininum example to reproduce the error. ``` import mxnet as mx from mxnet import gluon from mxnet.gluon import nn from mxnet.test_utils import assert_almost_equal from common import setup_module, with_seed import numpy as np import random from nose.tools import raises #mx.Context.default_ctx = mx.Context('gpu', 0) mx.Context.default_ctx = mx.Context('cpu', 0) layer = nn.Dense(1000) x = mx.nd.random.uniform(shape=(16, 128, 300, 300)) #x = mx.nd.random.uniform(shape=(1, 2, 30, 30)) x.attach_grad() layer.collect_params().initialize() with mx.autograd.record(): out = layer(x) out.backward() print x.grad.shape print x.grad.asnumpy().shape ``` Error message: print x.grad.asnumpy().shape File "/home/linliu/mxnet_gpu/incubator-mxnet/python/mxnet/ndarray/ndarray.py", line 1876, in asnumpy ctypes.c_size_t(data.size))) File "/home/linliu/mxnet_gpu/incubator-mxnet/python/mxnet/base.py", line 149, in check_call raise MXNetError(py_str(_LIB.MXGetLastError())) mxnet.base.MXNetError: ([04:25:06] include/mxnet/././tensor_blob.h:257: Check failed: this->shape_.Size() == shape.Size() (11520000000 vs. 2930065408) TBlob.get_with_shape: new and old shape do not match total elements
---------------------------------------------------------------- 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
