Kenneth-X opened a new issue #9344: fcn-xs  for mutil GPU
URL: https://github.com/apache/incubator-mxnet/issues/9344
 
 
   i follow the code:
   _python -u fcn_xs.py --model=fcn32s --prefix=VGG_FC_ILSVRC_16_layers 
--epoch=74 --init-type=vgg16_
   and change:
   _ctx = [mx.gpu(0),mx.gpu(1),mx.gpu(2),mx.gpu(3)]_  
   in 'fcn_x.py' file
   but i get the error:
   
    **File "/data-sdc/tao.xie/mxnet/python/mxnet/base.py", line 146, in 
check_call
       raise MXNetError(py_str(_LIB.MXGetLastError()))
   mxnet.base.MXNetError: [13:50:42] include/mxnet/./base.h:341: Invalid 
context string [gpu(0), gpu(1), gpu(2),gpu(3)]**
   
   then i debug it and find:  
   in 'init_fcnxs.py' file the function' init_from_vgg16'
   
   def init_from_vgg16(ctx, fcnxs_symbol, vgg16fc_args, vgg16fc_auxs):
       fcnxs_args = vgg16fc_args.copy()
       fcnxs_auxs = vgg16fc_auxs.copy()
       for k,v in fcnxs_args.items():
       if(v.context != ctx):
       **fcnxs_args[k] = mx.nd.zeros(v.shape, ctx)**
   
   it gets error in 'mx.nd.zeros(v.shape, ctx)', in which 'ctx = 
[mx.gpu(0),mx.gpu(1),mx.gpu(2),mx.gpu(3)] '
   
   it seems the ctx is Invalid ................
   
   how can i change it to train using mutil gpu? 
   need help?please
    
   

----------------------------------------------------------------
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