Skip to content

Conversation

@davidwyld
Copy link
Owner

@davidwyld davidwyld commented Jan 27, 2025

@MartinSJRogers - Tom and I are keen to move some of the patch stitching to the pred module from model since it operates on predictions.

Tests are passing but could you take a look a this please? In particular some of the docstrings and the descriptions of X_t to advise me of any changes as I'm not sure these make sense.

While we're at it, the docstring of _get_coordinate_extent I'm not really sure I follow:

Get coordinate extent of dataset. This method is applied to either X_t or patchwise predictions.

Could you suggest something more descriptive please?

Finally, should we also consider moving get_patch_overlap to the same module? And possibly moving its call inside stitch_clipped_predictions?

Copy link
Collaborator

@MartinSJRogers MartinSJRogers left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Approved- a few minor amendments to mark up text found in my PR: https://github.com/davidwilby/deepsensor/tree/mr_move_stitching_to_pred

Edit some markup text in new methods in pred module.
@davidwyld davidwyld merged commit c114926 into patchwise_train Feb 12, 2025
6 of 8 checks passed
@davidwyld davidwyld deleted the dw/move_stitching_to_predict branch February 12, 2025 10:29
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants