ModuleNotFoundError: No module named 'keras.legacy'

[20165645phungthitrang]

First of all, thanks for your repo! I am having problems importing the library, I tried to fix it but didn't fix it yet. Can you help me :(

[mrtushartiwari]

I am also facing the similar issue. Is there a solution for this apart from downgrading.

I tried to downgrade to !pip install tensorflow==2.2.0 !pip install keras==2.3.1 But later in my code keras is conflicting with tensorflow.

` import tensorflow as tf

import tensorflow.keras as keras

print(tf.version)

print(keras.version) ` Output: 2.4.1 2.4.0

!pip install -U segmentation-models==0.2.1

import segmentation_models as sm

Gives the following error: ModuleNotFoundError: No module named 'keras.legacy'

stack Trace: 
---------------------------------------------------------------------------

ModuleNotFoundError                       Traceback (most recent call last)

<ipython-input-16-7b5049dd4be0> in <module>()
----> 1 import segmentation_models as sm

6 frames

/usr/local/lib/python3.7/dist-packages/segmentation_models/__init__.py in <module>()
      4 
      5 from .unet import Unet
----> 6 from .fpn import FPN
      7 from .linknet import Linknet
      8 from .pspnet import PSPNet

/usr/local/lib/python3.7/dist-packages/segmentation_models/fpn/__init__.py in <module>()
----> 1 from .model import FPN
      2 

/usr/local/lib/python3.7/dist-packages/segmentation_models/fpn/model.py in <module>()
----> 1 from .builder import build_fpn
      2 from ..backbones import get_backbone, get_feature_layers
      3 from ..utils import freeze_model
      4 from ..utils import legacy_support
      5 

/usr/local/lib/python3.7/dist-packages/segmentation_models/fpn/builder.py in <module>()
      6 from keras.models import Model
      7 
----> 8 from .blocks import pyramid_block
      9 from ..common import ResizeImage
     10 from ..common import Conv2DBlock

/usr/local/lib/python3.7/dist-packages/segmentation_models/fpn/blocks.py in <module>()
      1 from keras.layers import Add
      2 
----> 3 from ..common import Conv2DBlock
      4 from ..common import ResizeImage
      5 from ..utils import to_tuple

/usr/local/lib/python3.7/dist-packages/segmentation_models/common/__init__.py in <module>()
      1 from .blocks import Conv2DBlock
----> 2 from .layers import ResizeImage

/usr/local/lib/python3.7/dist-packages/segmentation_models/common/layers.py in <module>()
      2 from keras.engine import InputSpec
      3 from keras.utils import conv_utils
----> 4 from keras.legacy import interfaces
      5 from keras.utils.generic_utils import get_custom_objects
      6 

ModuleNotFoundError: No module named 'keras.legacy'

---------------------------------------------------------------------------
NOTE: If your import is failing due to a missing package, you can
manually install dependencies using either !pip or !apt.

To view examples of installing some common dependencies, click the
"Open Examples" button below.
---------------------------------------------------------------------------

[sayakpaul]

It seems like TensorFlow 2 is not yet supported.

[sumanttyagi]

when i am running this line :from segmentation_models import Unet getting this error : AttributeError: module 'tensorflow_core.compat.v2' has no attribute 'internal'

can anybody help me with it i am using following env. !pip install segmentation-models !pip install tensorflow==2.1.0 !pip install keras==2.3.1

[sumanttyagi]

It seems like TensorFlow 2 is not yet supported.

can you please mention the dependency or requirement to run this repo... i have tried every combination though still not able to fix it...............

[gogitter24]

It seems like TensorFlow 2 is not yet supported.

can you please mention the dependency or requirement to run this repo... i have tried every combination though still not able to fix it...............

I'm currently using this repository with the following requirements: %tensorflow_version 1.x and downgrading the h5py as follows !pip install 'h5py==2.10.0' and then imported the repo as: !pip install -U segmentation-models obviously not tensorflow version 1.x isnt as convincing as 2.x ,but this isworking for me at the moment

[mrtushartiwari]

The following has worked for me on google colab as of now.

!pip3 install tensorflow==2.2
!pip3 install keras==2.3.1
!pip3 install -U segmentation-models

%env SM_FRAMEWORK=tf.keras
import segmentation_models as sm
tf.keras.backend.set_image_data_format('channels_last')