module 'keras.utils.generic_utils' has no attribute 'get_custom_objects' when importing segmentation_models

Question

I am working on google colab with the segmentation_models library. It worked perfectly the first week using it, but now it seems that I can't import the library anymore. Here is the error message, when I execute import segmentation_models as sm :

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

AttributeError                            Traceback (most recent call last)

<ipython-input-3-6f48ce46383f> in <module>
      1 import tensorflow as tf
----> 2 import segmentation_models as sm

            3 frames

/usr/local/lib/python3.8/dist-packages/efficientnet/__init__.py in init_keras_custom_objects()
     69     }
     70 
---> 71     keras.utils.generic_utils.get_custom_objects().update(custom_objects)
     72 
     73 

AttributeError: module 'keras.utils.generic_utils' has no attribute 'get_custom_objects'

Colab uses tensorflow version 2.11.0.

I did not find any information about this particular error message. Does anyone know where the problem may come from ?

Answer 1

谢谢，它有效，但我刚刚这样做了：

import os
os.environ["SM_FRAMEWORK"] = "tf.keras"

from tensorflow import keras
import segmentation_models as sm

Answer 2

尝试这个。有用。我在谷歌colab上查了一下

!pip install -U -q segmentation-models
!pip install -q tensorflow==2.2.1
!pip install -q keras==2.5
import os
os.environ["SM_FRAMEWORK"] = "tf.keras"

from tensorflow import keras
import segmentation_models as sm

Answer 3

有时遇到同样的问题。我是如何解决的：

1. 打开文件 keras.py，将所有 'init_keras_custom_objects' 更改为 'init_tfkeras_custom_objects'。

2. keras.py 的位置在错误消息中。在您的情况下，它应该位于 /usr/local/lib/python3.8/dist-packages/efficientnet/ 中