TypeError: _obtain_input_shape() got an unexpected keyword argument 'include_top'

[alyato]

@adamcasson when i call the resnet152, it show that
base_model = ResNet152(classes=8)

[ChristofHenkel]

change line 262 - 266 to:

input_shape = _obtain_input_shape(input_shape,
                                  default_size=img_size,
                                  min_size=197,
                                  data_format=K.image_data_format(),
                                  require_flatten=False)

Fixed it for me at least

[Ev]

Using TensorFlow backend.
Traceback (most recent call last):

File "", line 4, in
from resnet152 import ResNet152

File "resnet152.py", line 37, in
from keras.applications.imagenet_utils import _obtain_input_shape

ImportError: cannot import name _obtain_input_shape

[Ev]

what @ChristofHenkel said didn't fix the issue for me

[Ev]

switching from keras 2.2.2 to 2.0.0 didn't solve the issue for me

[Ev]

the initial code in this repo with tensorflow 2.0.0 works though. It is better to consider a downgrade on this case

[Eduardo]

My PR #2 fixes that issue.

[Adam Casson]

Merged PR from @edumucelli should fix