AttributeError: 'function' object has no attribute 'get_shape'

[Wyatt1994]

at line287 x = Lambda(lambda z: z[:, :, :, c * grouped_channels:(c + 1) * grouped_channels] if K.image_data_format() == 'channels_last' else lambda z: z[:, c * grouped_channels:(c + 1) * grouped_channels, :, :])(input)
this cannot be compiled.I try to define two functions to replace lambda,and then it works!

AttributeError: 'function' object has no attribute 'get_shape'` #5987

[titu1994]

What are the functions that you are defining ? There is no error for Tensorflow backend with TF din ordering. I have not yet set it up for Theano channels first dim ordering.

[Wyatt1994]

like this,https://github.com/fchollet/keras/issues/5987

[titu1994]

Hmm, i'll look into it. Its an issue with Lambda layers needing a shape predefined for them.

[cowboytong]

where can i find the weight of dpn training in imagenet?