Is model.predict() the solution ?

[athtareq]

Hello, how do you call the solution after the training is done ? Because when I use, say, model.compile(3) I get Cannot feed value of shape () for Tensor 'Placeholder_261:0', which has shape '(?, 1)'

Thank you

[lululxvi]

You can use Y = model.predict(X), where X is a N by d matrix.

[athtareq]

Thanks for the quick answer, but can you please give me a quick example ? Sorry and thanks !

[lululxvi]

Check the example: https://github.com/lululxvi/deepxde/blob/master/examples/Burgers.py

[athtareq]

But for one single point ? e.g y is the solution, how to write y(3) ? Again, sorry for the noobish questions

[lululxvi]

Then use X = np.array([[3]]).

[athtareq]

Thank you very much, you're a godsend !