R Devon Hjelm
R Devon Hjelm
Interesting. Can you do 2 things for me, as I cannot reproduce these errors: 1) provide the full traceback as when you run the script. 2) provide the value of...
Also, can you confirm your version of Theano?
Sorry, I knew about this issue and I need to modify the setup script to address this. In your ~/.theanorc (if it exists), you need the following fields: `[global]` `floatX=float32`....
I've added .theanorc to setup in the development branch and this will be merged soon.
Sorry for the late reply. Can you confirm your version of theano? thanks > On Sep 16, 2016, at 12:57 PM, karmeze [email protected] wrote: > > Hi Devon, > When...
I suggest you update to 0.9dev
Interesting, I wasn't aware of this distinction. Is there a backend solution that can manage this, or is it up to the user when they design routines?
So this was supposed to be default behavior at some point, but it was removed. That said, the way losses and results are handled in the backend are quite messy,...
OK! Looking forward to the PR!
Yes, this issue is still relevant. I'm not sure there's a way to fix it without modifying the pytorch data loader.