Keras as theano as backend can use the code of Keras as tensorflow backend?

1
votes

Can the code which is written on keras using tensorflow as backend can run on another keras environment were theano is the backend? Is there any computational benefits when running on any of them?

1
tensorflowdeep-learningkerastheano
It depends, if you are loading some weights computed by Tensorflow backend you can't run it in Theank backend. Anyways if you are not loading any weights you can do it, but remember to change the dimensions ordering. I think there is no computational difference between them, i'm not sure. - Eric
Yeah even I felt it. I tried to change the tensorflow to theano backend but there was some kind of dimension errors in the code as theano does'nt calculate the things automatically. Anyway I will find a fix for it and will post my further updates in the posts here. Thanks a lot for your reply. - Abin Jo Abraham

1 Answers

0
votes

To avoid dimensional errors just add

"image_data_format": "channels_last"

in the Keras.Json configuration file at

~\.keras

If there is not such a file you can create it yourself.