Python libraries being incompatible with each other - tensorflow

Question

I am trying to reproduce a piece of code source for Image Segmentation which involves the use of the
keras
and
segmentation-models
libraries.
I was able to run it in another PC running Python 3.9 and tensorflow==2.10.0, keras==2.10.0, segmentation-models==1.0.0. However, when I am trying to reproduce it on my PC with tensorflow==2.11.0, keras==2.11.0, and segmentation-models==1.0.1, I get this error:
File "/home/dev/anaconda3/lib/python3.9/site-packages/efficientnet/__init__.py", line 71, in init_keras_custom_objects
keras.utils.generic_utils.get_custom_objects().update(custom_objects)
AttributeError: module 'keras.utils.generic_utils' has no attribute 'get_custom_objects'
Library versions seem to be consistent with what is mentioned in requirements of source.
I tried downgrading keras, tensorflow, and segmentation-models versions to the older ones (which I reckoned had worked on the earlier PC), but that again shows up further compatibility issues. On further digging, I found a link which seemed to give the right requirements. I created a new conda environment with Python==3.7.0, and have again downgraded the versions into those suggested in link. I still am getting newer issues:
(sm) dev#Turbo:/media/dev/WinD/Curious Dev B/PROJECT STAGE - II$ python3 debug.py
Using TensorFlow backend.
Segmentation Models: using `keras` framework.
Traceback (most recent call last):
File "debug.py", line 150, in <module>
model = sm.Unet(BACKBONE, classes=n_classes, activation=activation)
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/segmentation_models/__init__.py", line 34, in wrapper
return func(*args, **kwargs)
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/segmentation_models/models/unet.py", line 226, in Unet
**kwargs,
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/segmentation_models/backbones/backbones_factory.py", line 103, in get_backbone
model = model_fn(*args, **kwargs)
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/classification_models/models_factory.py", line 78, in wrapper
return func(*args, **new_kwargs)
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/efficientnet/model.py", line 538, in EfficientNetB3
**kwargs)
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/efficientnet/model.py", line 368, in EfficientNet
img_input = layers.Input(shape=input_shape)
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/keras/engine/topology.py", line 1439, in Input
input_tensor=tensor)
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/keras/legacy/interfaces.py", line 87, in wrapper
return func(*args, **kwargs)
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/keras/engine/topology.py", line 1301, in __init__
name = prefix + '_' + str(K.get_uid(prefix))
File "/home/dev/anaconda3/envs/sm/lib/python3.7/site-packages/keras/backend/tensorflow_backend.py", line 63, in get_uid
graph = tf.get_default_graph()
AttributeError: module 'tensorflow' has no attribute 'get_default_graph'
I am unsure as to what went wrong here, the code worked six months ago. Now I am unable to run the same code, probably due to version changes of libraries. But at least it should have worked with the older versions that I myself used earlier, or at least with what link says.