Albert Zeyer

In `test_Data_get_common_data_extra2_static_spatial`, what happens when you do the same change, i.e. remove `auto_create_placeholders=True`?

The problem in `test_concat_new_dim_tag` (and probably others) is another bug in `Data.__init__`: When `auto_create_placeholders` is used, the size placeholders are created first, but then `same_dim_tags_as` comes afterwards.

Note there is already `Dim.is_dynamic` but a lot of code is checking `dimension is None` directly. All such code should use `is_dynamic` instead. Note that the `is_dynamic` implementation in principle...

We might need to introduce a new flag, specifically for this, like `_is_dynamic` or so.

Note, I'm already continually fixing any `dimension is None` checks. Note that those checks don't necessarily always map to `is_dynamic`. They sometimes also check whether the dim is defined/known yet....

We also need to be careful in `get_for_batch_ctx`. In the case the dim is not yet defined, this should also not do anything. Otherwise a static dim cannot properly be...

I'm thinking about whether this automatic introduction of implicit dims was a good idea. This original code really should work, as it would be quite confusing otherwise: ```python x =...

@Zettelkasten what do you think about this? Specifically, here (https://github.com/rwth-i6/returnn/issues/706#issuecomment-964379170) you said: > Do we want to force the user to specify implicit dims if they specify `out_shape`? I tend...

We could also allow returnn-common to be a bit more relaxed about this and `verify_out_shape` on returnn-common side would allow to ignore implicit dims, or maybe have an option `check_implicit_dims`...

> We could also allow returnn-common to be a bit more relaxed about this and `verify_out_shape` on returnn-common side would allow to ignore implicit dims, or maybe have an option...