Error while import keras: AttributeError: module 'tensorflow.compat.v2.__internal__' has no attribute 'register_clear_session_function'
Question:
I want to import keras
after I did pip install keras
, but it shows message as shown below. I even can’t call any function from keras library. Can anyone know about this?
import keras
Error:
AttributeError: module 'tensorflow.compat.v2.__internal__' has no attribute 'register_clear_session_function'
Answers:
after tf2, you should use
from tensorflow import keras
!pip install -U -q segmentation-models
!pip install -q tensorflow==2.1
!pip install -q keras==2.3.1
!pip install -q tensorflow-estimator==2.1.
## Imports libs
import os
os.environ['CUDA_VISIBLE_DEVICES'] = '0'
os.environ["SM_FRAMEWORK"] = "tf.keras"
from tensorflow import keras
import segmentation_models as sm
you can should use import tensorflow.keras
instead of import keras
.
More info here
This can be caused by a version mismatch between Keras installation and Tensorflow. Make sure their versions match.
None of the other answers here worked for me. I ended up updating keras and tensorflow as follows:
pip install --upgrade keras
pip install --upgrade tensorflow
Now, it works for me.
from keras.models import Model
ERROR:
AttributeError: module ‘tensorflow.compat.v2.internal‘ has no attribute ‘register_clear_session_function’
I want to import keras
after I did pip install keras
, but it shows message as shown below. I even can’t call any function from keras library. Can anyone know about this?
import keras
Error:
AttributeError: module 'tensorflow.compat.v2.__internal__' has no attribute 'register_clear_session_function'
after tf2, you should use
from tensorflow import keras
!pip install -U -q segmentation-models
!pip install -q tensorflow==2.1
!pip install -q keras==2.3.1
!pip install -q tensorflow-estimator==2.1.
## Imports libs
import os
os.environ['CUDA_VISIBLE_DEVICES'] = '0'
os.environ["SM_FRAMEWORK"] = "tf.keras"
from tensorflow import keras
import segmentation_models as sm
you can should use import tensorflow.keras
instead of import keras
.
More info here
This can be caused by a version mismatch between Keras installation and Tensorflow. Make sure their versions match.
None of the other answers here worked for me. I ended up updating keras and tensorflow as follows:
pip install --upgrade keras
pip install --upgrade tensorflow
Now, it works for me.
from keras.models import Model
ERROR:
AttributeError: module ‘tensorflow.compat.v2.internal‘ has no attribute ‘register_clear_session_function’