Skip to content

Remove event model part 2#966

Draft
jwaiton wants to merge 55 commits into
next-exp:masterfrom
jwaiton:remove-evm-2
Draft

Remove event model part 2#966
jwaiton wants to merge 55 commits into
next-exp:masterfrom
jwaiton:remove-evm-2

Conversation

@jwaiton

@jwaiton jwaiton commented Jun 19, 2026

Copy link
Copy Markdown
Member

This is (hopefully) one of the last steps in removing the event model, which relied on the refactoring of Paolina.

In this (overly large PR):

  • Removal of MCHit(), MCInfo(), Waveform(), KrEvent(), Hit(), HitCollection(), VoxelCollection(), Track() classes and related functionality,
  • inclusion of the hit_type, kr_event_type dictionaries,
  • altering paolina functions to work with the new dataframe-based structure
  • altering of paolina tests to work with the new dataframe-based structure

This is a massive PR, with lots changing all at once. The method for which we tag blobs exists both in the old fashioned way (outputting a dataframe table with the topological information) and in the new way (having the hits labelled as being part of each blob).

Voxels now have ID hashes, and the voxel's hits now have voxel_id within their dataframe to relate them.

There are even more minor changes that will require more thorough of a review to re-spot.

One missing feature is the method of moving the blob centre to encapsulate the highest energy voxels. I will implement this promptly.

gonzaponte and others added 30 commits June 19, 2026 18:25
This reverts commit 3a8e708a38ac0f10482c933d2f9b609feab5cfd1.
qmax was not present in the list of columns, as the files were created
before the column was introduced
Here we lose the blob radius/scan_radius methodology for
the blob centre. This will be reintroduced in a later commit
Removing `scan_radius` as a parameter
- include `get_track_energy()`
- pass through `voxel_size` into `drop_voxels()` &
  `drop_voxel_inplace()`
- minor bug fixes across voxel dropping functions
This related to a blob method that I've removed, as I don't want every
hit to have a new element written to it. Instead the hit-blob
relation stays as it was previously; calculated in-situ.
jwaiton added 15 commits June 19, 2026 18:36
- fixed voxels note being modified in place due to use of copy
- fixed dropped hits removed on incorrect voxel index
- fixed energy type mapping across `make_tracks()`
`test_tracks_with_dropped_voxels()` needed altering as prior test expected `strict_voxel_size = False`, which resulted in a differing number of voxels being dropped.
`test_initial_voxels_are_the_same_after_dropping_voxels()` no longer
makes sense due to the in-place nature of `drop_voxels()`, so it has
been modified to have the same spirit, becoming
`test_initial_voxels_are_modified_inplace_after_dropping_voxels()`
We no longer have the capability of voxels without hits, so this test no
longer makes sense.
`test_paolina_functions_with_hit_energy_different_from_default_value()`
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants