Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Theano? #7

Open
riturkki opened this issue Feb 2, 2017 · 5 comments
Open

Theano? #7

riturkki opened this issue Feb 2, 2017 · 5 comments

Comments

@riturkki
Copy link

riturkki commented Feb 2, 2017

Thanks for the great benchmark!

Are you considering to include Theano in the future?

@shyhuai
Copy link
Collaborator

shyhuai commented Feb 4, 2017

It may be great if there are more tools included. We will include some other existing popular tools for comparison in the future, but it may needs some time to update with more tools since we need to be very careful to make sure the networks' settings are all the same in different tools.

@botev
Copy link

botev commented Feb 27, 2017

Do you mind if I try to build the examples for Theano (single GPU only) and submit a PR?

@shyhuai
Copy link
Collaborator

shyhuai commented Feb 27, 2017

That's good. Thanks for contributing to the codes of Theano examples.

@botev
Copy link

botev commented Feb 27, 2017

Could you however elaborate on the setup you have. From some of the code it seems you are using pytho2.7 rather than 3 and also virtual environemnts. Without the exact details it is quite hard to reproduce the benchmark on our own systems as well as just testing that integrating other frameworks would work

@FreemanX
Copy link
Collaborator

We are working on our documentation, and soon we will release a more detailed instructions for adding new tools. As for your specific questions:

  1. We are using python2.7 for our benchmark framework as well as deep learing tools that support python.
  2. The virtual environment was designed for old version of tensorflow. When we started this project, CUDA8.0 was just released and tensorflow didn't support it at that time. So we created an environment containing CUDA7.5 just for tensorflow. Now it's not necessary any more.
    If you encounter any questions, please don't hesitate to ask.

Thank you for your efforts!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants