Loss function exploding with a different data set

Course Q&A TensorFlow Developer Professional Introduction to TF for Artificial Intelligence ...
1

Hi,

Loss function is exploding with the dataset:

xs = np.array([1, 2, 10, 11, 20, 21], dtype=float)
ys = np.array([1.0, 1.5, 5.5, 6.0, 10.5, 11.0], dtype=float)

However, with the suggested dataset, the code is working correctly.

xs = np.array([1, 2, 3, 4, 5, 6], dtype=float)
ys = np.array([1.0, 1.5, 2.0, 2.5, 3.0, 3.5], dtype=float)

I am curious why is this happening.

Thank you

2

Can you please send the notebook you are using as pdf so that I can have a look at the loss function you are using and also the learning rate
I think its the learning rate parameter that is causing the loss to go haywire
To confirm I would need the code
Thanks