ImportError: cannot import name 'normalize_data_format'

[zeeshannisar]

I have read your article and its pretty nice enough to understand. Thank you for your contribution. When I a trying to train at my own using your code it gives me an Import Error like following. I am using google Colab environment. Having some search over the error i got that the following line is compatible to keras version==2.2.2. I have also installed that yet not solved with the error. Please help me to get over it. By default keras version installed in colab is 2.2.4

---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
<ipython-input-47-f8ce7e15cf87> in <module>()
      9 from keras.layers.merge import Add
     10 from keras.utils import conv_utils
---> 11 from keras.utils.conv_utils import normalize_data_format
     12 
     13 from keras.layers.core import Dropout

ImportError: cannot import name 'normalize_data_format'

---------------------------------------------------------------------------

[ruixv]

Hi, I met the same question and solved it using the method here: keras-team/keras-contrib#298 (comment)

#from keras.backend.common import normalize_data_format
from keras.utils.conv_utils import normalize_data_format

that is:
self.data_format= common.normalize_data_format(data_format)

[zhangxiaoya]

I used keras 2.1.3 and tensorflow 1.5 and solved it.

[gdyshi]

In the new version of keras, this function was not put in conv_utils, but in the backend.
I sloved by

import keras.backend as K
and replace "conv_utils.normalize_data_format" with "K.normalize_data_format"

[docongminh]

I used keras 2.1.3 and tensorflow 1.5 and solved it.
thanks ! It work

[sailfish009]

keras version: 2.4.3
K.common.normalize_data_format(data_format)
AttributeError: module 'keras.backend' has no attribute 'common'

[sailfish009]

https://stackoverflow.com/questions/53442190/importerror-cannot-import-name-normalize-data-format

import keras.backend as K
def normalize_data_format(value):
    if value is None:
        value = K.image_data_format()
    data_format = value.lower()
    if data_format not in {'channels_first', 'channels_last'}:
        raise ValueError('The `data_format` argument must be one of '
                         '"channels_first", "channels_last". Received: ' +
                         str(value))
    return data_format

[superran1]

I meet same problem with you ,keras version is 2.4.3.

[gevindharmalim]

can someone tell me how to solve it in keras version 2.4.3