Open haniehkh18 opened 5 years ago
Hi,
Have you tried this solution? $HOME/.keras/keras.json as vim /home/iot/.keras/keras.json change last "backend": "tensorflow" to "theano"
Hello Longxuan
Thanks for your reply. I have already set the backend to theano.
Regards; Hanieh
On Sun, May 19, 2019 at 11:44 AM malongxuan notifications@github.com wrote:
Hi,
Have you tried this solution? $HOME/.keras/keras.json as vim /home/iot/.keras/keras.json change last "backend": "tensorflow" to "theano"
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/malongxuan/QAselection/issues/2?email_source=notifications&email_token=AKZM5NWHA2V2M7XCW7CEXIDPWD46HA5CNFSM4HNLNABKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVW4KTI#issuecomment-493733197, or mute the thread https://github.com/notifications/unsubscribe-auth/AKZM5NROMXT2SJVNPO4FFBLPWD46HANCNFSM4HNLNABA .
Hello
I get the error: "'module' object has no attribute 'T' " on this part of the code: k_max_e = K.T.set_subtensor(g[K.T.arange(g.shape[0]).dimshuffle(0,'x','x'), K.T.arange(g.shape[1]).dimshuffle('x',0,'x'), K.T.argsort(g)[:,:,:bound]], 0.0) all requirement are already installed.