The code in the notebook appears to be malfunctioning. while trying to train the model (strictly following the instructions in the class documentation) the loss function outputs np.nan is all training epochs. For that reason, the prediction output of the model after training is continuous, impacting the final MAE - which is controlled to pass the assignment.
I suspect that there are issues with the TensorFlow installation, which are causing parameter updates to fail and resulting in missing values during training.
Am I the only one noticing this issue? Am I missing something?
can you please screenshot of the issue you are encountering, so we have a better understanding of your issue. Please make sure not to post any grade cell codes as it is against community guidelines.
Regards
DP