training problem probably from model construction

[Ting-Bao]

When I train the model with your code main_semi.py, after some epochs, I got the error from VAE.fit():

Exception has occurred: InvalidArgumentError
2 root error(s) found.
(0) INVALID_ARGUMENT: slice index 0 of dimension 0 out of bounds.
[[{{node lambda/strided_slice_3}}]]
[[loss/mul/_241]]
(1) INVALID_ARGUMENT: slice index 0 of dimension 0 out of bounds.
[[{{node lambda/strided_slice_3}}]]

It seems that the some dimension of the data or model gets wrong, strangely the training can be excuted for over 100 epochs before raise error. I use:
max_elms = 4
min_elms = 3
max_sites = 8

[dannyzekunren]

Hi Ting-Bao. It seems to be the slicing problem during the training. We will work on it. One workaround would be to increase the batch size. Can you try to double the batch size?

[Ting-Bao]

Yes, larger batch size may avoid the problem.
We suspect the slicing error occurs when no material has a semi property (eg, powerfactor) tag in a specific batch.
In extreme, use batchsize=1 and the error will be raised in the first epoch.