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

Reply via email to