Training error about anomalydetectioncvpr2018

[draxler1]

Any solution to this??

[mvk07]

Please use theano backend. This is due to tensorflow vector being passed to theano flatten.

Use this link for more information: https://keras.io/backend/

[NitinDatta8]

I solved the problem using

import os
os.environ['KERAS_BACKEND'] = 'theano'

[Mohamed-Salah62]

I tried it and it didn't work

I solved the problem using

import os
os.environ['KERAS_BACKEND'] = 'theano'

[mvk07]

I tried it and it didn't work

I solved the problem using

import os
os.environ['KERAS_BACKEND'] = 'theano'

could you post the error you encounter

[Mohamed-Salah62]

AttributeError: 'Tensor' object has no attribute 'reshape'

import os
os.environ['KERAS_BACKEND'] = 'theano'

could you post the error you encounter

[yassinesamet]

@mvk07
@NitinDatta8
@draxler1
@Mohamed-Salah62
@qwdr7777
any solution pleaaase

import os
os.environ['KERAS_BACKEND'] = 'theano'

----------------------------->I tried it and it didn't work