[theano-users] Could not initialize pygpu, support disabled

I am trying to test theano with GPU (AMD graphics card). I have installed latest versions of theano and libgpuarray. However, I get the following error when trying to execute test script from: http://deeplearning.net/software/theano/tutorial/using_gpu.html THEANO_FLAGS=mode=FAST_RUN,device=opencl

[theano-users] Failed to infer_shape from Op Images2Neibs{valid}.

Hi, I am getting the following error when using images2neibs with neib_shape is given as variables and used in a function with scan. python shapebug.py Using gpu device 0: Quadro K4200 (CNMeM is disabled, cuDNN 5004) theano version 0.8.2 WARNING (theano.tensor.opt): Failed to infer_shape from O

[theano-users] 'module' object has no attribute 'Subtensor'

Hello, I would like to submit this puzzle to the Theano-users-group: 3 years ago we used the cPickle module to store some convolutional network data that had been produced using Theano (using the class called " LeNetConvPoolLayer"). Here is the error we get when we now try to unpickle this data,

Re: [theano-users] 'module' object has no attribute 'Subtensor'

If you use the same Theano version, it should work. Can you remove all Theano version. Try to import it and make sure it fail. Then install the old version. If you want to try with a recent Theano, give me the full error message. I could send you a diff of how to fix it. It is not sure we will sup

Re: [theano-users] 'module' object has no attribute 'Subtensor'

Merci bien Frédéric pour cette réponse si rapide (je continue en français sauf contre-ordre, on parle français aussi!). Il est possible en effet qu'on n'utilise pas la bonne version de Theano, je vais essayer de la retrouver. Stéphane Dufau (en copie) voudra peut-être suivre ta proposition et essa

Re: [theano-users] 'module' object has no attribute 'Subtensor'

Pas de contre indication pour parlé Français, sauf que cela va diminuer énormément ceux qui peuvent comprendre la solution s'ils ont le même problème. Fred On Wed, Sep 21, 2016 at 11:04 AM, Thomas Hannagan wrote: > Merci bien Frédéric pour cette réponse si rapide (je continue en français > sauf

[theano-users] Dropout in Scan

Hi, I'm implementing a variation of a Pointer Network, and I'd like to add dropout to the hidden layers, but I'm having trouble using the random number generator. Details are in this post: http://stackoverflow.com/questions/39606372/dropout-in-scan-theano Thank You, Peter -- --- You receiv

Re: [theano-users] Re: how to check variable's shape in theano.function

On Tue, Sep 20, 2016, GUKBEOM LEE wrote: > nouiz, thank you for advice. > But I don't understand what turbo function means. Probably "theano" has been autocorrected to "turbo". > Also, do real days mean > 'numpy.ndarray' instead of theano symbolic variable? Yes. > I'll try it. I appreciate yo

Re: [theano-users] Failed to infer_shape from Op Images2Neibs{valid}.

On Wed, Sep 21, 2016, Petar Palasek wrote: > Hi, > > I am getting the following error when using images2neibs with neib_shape is > given as variables and used in a function with scan. > > python shapebug.py > Using gpu device 0: Quadro K4200 (CNMeM is disabled, cuDNN 5004) > theano version 0.8.

Re: [theano-users] Re: how to check variable's shape in theano.function

Typo. Real days, should be real values like ndarray Turbo should be Theano. Le 20 sept. 2016 22:10, "GUKBEOM LEE" a écrit : > nouiz, thank you for advice. > But I don't understand what turbo function means. Also, do real days mean > 'numpy.ndarray' instead of theano symbolic variable? > I'll t

[theano-users] Error when running theano.test()

I just installed theano and I ran theano.test() to check to see if it was installed correctly. However, I keep on getting this error when running each test: Problem occurred during compilation with the command line below: C:\MinGW\bin\g++.exe -shared -g -O3 -fno-math-errno -Wno-unused-label -W