Unable to use theano backend

I installed the anaconda distribution with python 2.7 and the rest of the packages using conda as the package manager. I am using cuda 8.0. Here is how my theanorc looks like:

floatX = float32
device = cuda0

fastmath = True


Can someone please help?


Try conda install theano pygpu

Didn’t help. I will try using cuda 8.0 instead and see what happens.

Okay, installing cuda 8.0 also didn’t help…

I also have CUDA8, and I also cannot get Theano to work as backend, even though I get another error, which is unrelated to yours. Theano doesn’t support CUDA8.

Off-topic, but heads up about Theano:

September 28, 2017:

Dear users and developers,

After almost ten years of development, we have the regret to announce
that we will put an end to our Theano development after the 1.0 release,
which is due in the next few weeks. We will continue minimal maintenance
to keep it working for one year, but we will stop actively implementing
new features. Theano will continue to be available afterwards, as per
our engagement towards open source software, but MILA does not commit to
spend time on maintenance or support after that time frame.
– Yoshua Bengio,
Head of MILA


Thank you for your replies! If I have to run the notebooks on my own machine (which has a GPU), can someone please suggest how to set it up. Do the notebooks work with Tensorflow as the backend for Keras?

Install tensorflow’s GPU version and have a check on .theanoric (config file) in your home directory…for theano as backend…