By running the copy of the colab I have the following error. on my computer it runs without error

[dimkiriakos]

/usr/local/lib/python3.7/dist-packages/tensorflow/python/keras/optimizer_v2/optimizer_v2.py:375: UserWarning: The lr argument is deprecated, use learning_rate instead.
"The lr argument is deprecated, use learning_rate instead.")

ValueError Traceback (most recent call last)
in ()
2 epochs = 15
3
----> 4 model.compile(loss="categorical_crossentropy", optimizer=MadGrad(lr=1e-3), metrics=["accuracy"])
5 model.fit(x_train, y_train, batch_size=batch_size, epochs=epochs, validation_split=0.1)

5 frames
/usr/local/lib/python3.7/dist-packages/keras/optimizers.py in get(identifier)
131 else:
132 raise ValueError(
--> 133 'Could not interpret optimizer identifier: {}'.format(identifier))

ValueError: Could not interpret optimizer identifier: <madgrad.madgrad.MadGrad object at 0x7f8690765110>

[DarshanDeshpande]

I am able to reproduce this on Colab. Seems more like TF 2.6.0 is the issue. It works fine with TF <=2.5.
I am a bit busy at the moment to look into this so if you are willing to contribute and find a fix then please let me know.