[theano-users] How to partially override gradient computation

Asked as here . Encountered the problem when I'm trying to implement an approximation method for Real-Time-Recurrent-Learning. -- --- You received this message because you are subscribed to

RE: [theano-users] TypeError: Cannot convert Type TensorType(float64, 3D) (of Variable Subtensor{:int64:}.0) into Type TensorType(float64, (False, True, False)). You can try to manually convert Subten

Hi Fred, What command would I use to uninstall theano. I also installed the latest development version.Does it need a different command to uninstall? Thanks. Chi From: theano-users@googlegroups.com [mailto:theano-users@googlegroups.com] On Behalf Of Frédéric Bastien Sent:

Re: [theano-users] Pooling ignore_border = False pulls things off GPU ?

This is a know problem. We have prs that should fix that. I think you can just use them for now. I didn't see any real benefit of ignore_border=False. If you find some, tell me. But cudnn don't support that, so it will always be faster to set it to True. There is an issue about this, but I'm

Re: [theano-users] TypeError: Cannot convert Type TensorType(float64, 3D) (of Variable Subtensor{:int64:}.0) into Type TensorType(float64, (False, True, False)). You can try to manually convert Subten

To be safer, uninstall Theano a few time before and make sure you can't import it before reinstalling. Le 15 nov. 2016 16:20, "Pascal Lamblin" a écrit : > On Tue, Nov 15, 2016, Chi Ku wrote: > > Hi Pascal, > > > >How would I find out which version of Theano I

RE: [theano-users] TypeError: Cannot convert Type TensorType(float64, 3D) (of Variable Subtensor{:int64:}.0) into Type TensorType(float64, (False, True, False)). You can try to manually convert Subten

Hi Pascal, I did have 0.8.2. I installed the latest development version. But, I got the same error. I tried changing my program in various way to work around this without any success. Could you recommend a work around? Thanks. Chi -Original Message- From:

Re: [theano-users] TypeError: Cannot convert Type TensorType(float64, 3D) (of Variable Subtensor{:int64:}.0) into Type TensorType(float64, (False, True, False)). You can try to manually convert Subten

On Tue, Nov 15, 2016, Chi Ku wrote: > Hi Pascal, > >How would I find out which version of Theano I have? You can try to print theano.version.version, that should be generated when you call "pip install" >Can I use the following command to install the latest development version? > >

Re: [theano-users] core dump error

Use the theano flag: linker=py This will make things much slower, but it should raise a python error instead of a core dump. On Tue, Nov 15, 2016 at 10:00 AM, H wrote: > So I moved to another virtual environment which I use only development > version of Theano. I still

RE: [theano-users] TypeError: Cannot convert Type TensorType(float64, 3D) (of Variable Subtensor{:int64:}.0) into Type TensorType(float64, (False, True, False)). You can try to manually convert Subten

Hi Pascal, How would I find out which version of Theano I have? Can I use the following command to install the latest development version? pip install <--user> <--no-deps> git+https://github.com/Theano/Theano Thanks. Chi -Original Message- From:

Re: [theano-users] core dump error

So I moved to another virtual environment which I use only development version of Theano. I still get the same error without any more information. How can I figure out where did I do division by zero? -- --- You received this message because you are subscribed to the Google Groups

Re: [theano-users] core dump error

Actually I have installed a version using Bleeding-edge installation and I also have anaconda. How can I figure out which one I'm using? On Tuesday, November 15, 2016 at 2:47:44 PM UTC+1, Pascal Lamblin wrote: > > The usual immediate cause is an integer division by 0. > > This is sometimes

Re: [theano-users] core dump error

The usual immediate cause is an integer division by 0. This is sometimes introduced by trying to do shape inference when one of the dimensions of a tensor has a length of 0. I think we fixed one recently, are you using the development version? On Tue, Nov 15, 2016, H wrote: > Hey guys, > > I

[theano-users] core dump error

Hey guys, I got this error when I tried to run my gpu based script. It used to work fine before. Do you know what's happening? Floating point exception (core dumped) thanks, -- --- You received this message because you are subscribed to the Google Groups "theano-users" group. To