Re: [theano-users] Re: theao install nvcc fatal : Unknown option 'fPIC'

2017-03-21 Thread 李奕 
when I comment the compile_dir in the theanrc.txt, the error miss;

在 2017年3月16日星期四 UTC+8下午10:47:49，李奕写道：
>
> I try the command line but the error exists;  the trick is confusing; I 
> install theano multiple times. This's first time to have this error, May 
> the cuda not config properly? But I can run theano before and run 
> tensorflow on this server.
>
> 在 2017年3月14日星期二 UTC+8下午10:22:26，nouiz写道：
>>
>> The command is: conda install -c rdonnelly theano
>>
>> On Mon, Mar 13, 2017 at 5:42 PM Frédéric Bastien  
>> wrote:
>>
>>> Try Theano beta conda package
>>>
>>> Conda install -r raydonelly Theano
>>>
>>> From memory, I'm offline. If that don't work, look in Theano issues I 
>>> give the good commande line.
>>>
>>> Fred
>>>
>>> Le ven. 10 mars 2017 22:01, 李奕  a écrit :
>>>
 Thanks, I use anaconda, So what to do next to slove this problem ?

 在 2017年3月3日星期五 UTC+8下午9:46:18，nouiz写道：
>
> How did you get Python? If you compiled it yourself, you missed a 
> compilation option. I have a stack overflow answer to that. Search for 
> fPIC 
> and Theano.
>
> Fred
>
> Le mar. 28 févr. 2017 23:23, 李奕  a écrit :
>
 Thanks, very much. Actually In my anaconda version, theano + 
>> gpu(cuda8.0) + python3.5 is OK, I want to install in virtual 
>> environments.
>> In virtual envirment, I run "pip install --upgrade --no-deps 
>> git+git://github.com/Theano/Theano.git" to install the newest 
>> version;
>> But when I run import theano, the errors(unknown option 'fPIC') 
>> exists. 
>> My environment is conda(virtualenv)'s virtual environment python2, So 
>> I wonder whether the virtual environments has effect on theano? Thanks.
>>
>> 在 2017年2月28日星期二 UTC+8下午9:47:41，Ankit Shah写道：
>>
>>> This is an error related to conversion of floating point number 
>>> approximation. Use latest version of theano and keras [if using] there 
>>> is a 
>>> fix provided 
>>>
>>> On Tuesday, February 28, 2017 at 11:55:06 AM UTC+5:30, 李奕 wrote:

 Hello, 

 when I run import theano, the error is:
 ['nvcc', '-shared', '-O3', '-use_fast_math', '--compiler-bindir', 
 '/usr/local/cuda/bin/nvcc', '-m64', '-Xcompiler', 
 '-DCUDA_NDARRAY_CUH=c72d035fdf91890f3b36710688069b2e,-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION,-fPIC,-fvisibility=hidden',
  
 '-Xlinker', 
 '-rpath,/home/mlx/.theano/compiledir_Linux-4.2--generic-x86_64-with-debian-jessie-sid-x86_64-2.7.13-64/cuda_ndarray',
  
 '-I/home/mlx/anaconda3/envs/py2/lib/python2.7/site-packages/theano/sandbox/cuda',
  
 '-I/home/mlx/anaconda3/envs/py2/lib/python2.7/site-packages/numpy/core/include',
  
 '-I/home/mlx/anaconda3/envs/py2/include/python2.7', 
 '-I/home/mlx/anaconda3/envs/py2/lib/python2.7/site-packages/theano/gof',
  
 '-L/home/mlx/anaconda3/envs/py2/lib', '-o', 
 '/home/mlx/.theano/compiledir_Linux-4.2--generic-x86_64-with-debian-jessie-sid-x86_64-2.7.13-64/cuda_ndarray/cuda_ndarray.so',
  
 'mod.cu', '-lcublas', '-lpython2.7', '-lcudart']
 ERROR (theano.sandbox.cuda): Failed to compile cuda_ndarray.cu: 
 ('nvcc return status', 1, 'for cmd', 'nvcc -shared -O3 -use_fast_math 
 --compiler-bindir /usr/local/cuda/bin/nvcc -m64 -Xcompiler 
 -DCUDA_NDARRAY_CUH=c72d035fdf91890f3b36710688069b2e,-DNPY_NO_DEPRECATED_API=NPY_1_7_API_VERSION,-fPIC,-fvisibility=hidden
  
 -Xlinker 
 -rpath,/home/mlx/.theano/compiledir_Linux-4.2--generic-x86_64-with-debian-jessie-sid-x86_64-2.7.13-64/cuda_ndarray
  
 -I/home/mlx/anaconda3/envs/py2/lib/python2.7/site-packages/theano/sandbox/cuda
  
 -I/home/mlx/anaconda3/envs/py2/lib/python2.7/site-packages/numpy/core/include
  
 -I/home/mlx/anaconda3/envs/py2/include/python2.7 
 -I/home/mlx/anaconda3/envs/py2/lib/python2.7/site-packages/theano/gof 
 -L/home/mlx/anaconda3/envs/py2/lib -o 
 /home/mlx/.theano/compiledir_Linux-4.2--generic-x86_64-with-debian-jessie-sid-x86_64-2.7.13-64/cuda_ndarray/cuda_ndarray.so
  
 mod.cu -lcublas -lpython2.7 -lcudart')
 WARNING (theano.sandbox.cuda): The cuda backend is deprecated and 
 will be removed in the next release (v0.10).  Please switch to the 
 gpuarray 
 backend. You can get more information about how to switch at this URL:
  
 https://github.com/Theano/Theano/wiki/Converting-to-the-new-gpu-back-end%28gpuarray%29

 WARNING (theano.sandbox.cuda): CUDA is installed, but device gpu is 
 not available  (error: cuda unavailable)


 So what's the going on here? Thanks very much

>>> -- 
>>
>> --- 
>> You received this

[theano-users] Re: about conv2d on CPU

2017-03-21 Thread Jesse Livezey 
That is correct as of theano 0.8 (I think).

If you use the bleeding edge version of theano, you can let CorrMM use 
openmp to parallelize across batches. If you have more than 2 cores, this 
should give additional speedup. GPUs are going to be much faster than CPUs 
generally, if you have large batches and lots of cores, CPUs can catch up a 
bit, but GPUs are still going to be faster.

On Monday, March 20, 2017 at 11:59:52 PM UTC-7, C. Ng wrote:
>
> Hi,
>
> Just want to confirm that theano.tensor.nnet.conv2d uses CorrMM (not the 
> legacy convolution) by default in CPU mode ?
>
> I was hoping that forward prop (doing inference only, no training) using 
> CPU for convolution might be as fast as GPU (using CorrMM), given my batch 
> size is only 10. But using GPU is still quite a bit faster.  
>
>
>
>
>  
>

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"theano-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to theano-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[theano-users] about conv2d on CPU

2017-03-21 Thread C. Ng 
Hi,

Just want to confirm that theano.tensor.nnet.conv2d uses CorrMM (not the 
legacy convolution) by default in CPU mode ?

I was hoping that forward prop (doing inference only, no training) using 
CPU for convolution might be as fast as GPU (using CorrMM), given my batch 
size is only 10. But using GPU is still quite a bit faster.  




 

-- 

--- 
You received this message because you are subscribed to the Google Groups 
"theano-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to theano-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.