Conversation
Discontinuity built its mask with jnp.ones_like(x), where x is the grid of shape (D, ..., N). The mask therefore kept D channels, so RandomDiscontinuities returned D identical channels instead of the (1, ..., N) shape every other IC generator returns. Build the mask from a single grid slice and add RandomDiscontinuities to the output shape test.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Discontinuity.__call__built its mask withjnp.ones_like(x, dtype=bool), wherexis the grid of shape(D, ..., N). The mask therefore keptDchannels, so in 2d and 3dRandomDiscontinuitiesreturnedDidentical channels instead of the(1, ..., N)shape the other IC generators return:This PR builds the mask from a single grid slice (
x[0:1]). 1d output is unchanged.RandomDiscontinuitieswas also missing from the parametrization oftest_ic_output_shape, so I added it there. Its 2d and 3d cases fail onmainand pass with this change, andpytest tests/test_ic.py -k Discontinuitiespasses (20 tests). The only usage in the docs is the 1d showcase notebook, which is not affected.I found this while checking
build_ic_setagainst every random IC generator for #58.