Theano development will stop after release of version 1.0 in a few weeks: https://groups.google.com/forum/#!topic/theano-users/7Poq8BZutbY … - end of an era...
-
Show this thread
-
Theano has had a huge impact on the machine learning tooling landscape. It has pioneered a lot of the ideas that we take for granted today.
1 reply 25 retweets 108 likesShow this thread -
Big respect to the Theano developers and contributors for the years of hard work and the impact achieved. Theano is still used by many.
4 replies 21 retweets 140 likesShow this thread -
Replying to @fchollet
How hard was it to wrap keras around both? Do you have a list of things to do if you want to move a project to support both?
1 reply 0 retweets 0 likes -
Both being TF and Theano...
1 reply 0 retweets 0 likes -
Replying to @dmarthal
For a project to support all backends, you just need to make sure that you don't call TF-specific (of Theano-specific) functions
1 reply 0 retweets 1 like
I.e. that you only rely on public Keras features or keras.backend function calls
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.