Import error for Keras pre trained models in monk

[Ayush6459]

hii , I am trying to use monk for Keras prototype for pre-trained image classification models like resent50 but the backend is giving me error like this :

ImportError Traceback (most recent call last)
in ()
----> 1 ptf.Default(dataset_path=["./dataset/train", "./dataset/val"], model_name="resnet50", freeze_base_network=True, num_epochs=5);

11 frames
/usr/local/lib/python3.7/dist-packages/monk/tf_keras_1/models/models.py in get_base_model(model_name, use_pretrained, num_classes, freeze_base_network, input_size)
62 from keras.applications import NASNetLarge as keras_model
63 elif(model_name == "resnet50"):
---> 64 from keras.applications import ResNet50 as keras_model
65 elif(model_name == "resnet101"):
66 from keras.applications import ResNet101 as keras_model

ImportError: cannot import name 'ResNet50' from 'keras.applications' (/usr/local/lib/python3.7/dist-packages/keras/applications/init.py)

---

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

I am using monk in Colab , Please see the issue .

[ravikanur]

This is basically a version issue. In the current version keras has been included inside tensorflow. Hence while importing it should be imported as
from tensorflow.keras.applications import ResNet50

[Chaituvarma-109]

Having the same issue. How to solve it