Make CKF propagation a 1:1 kernel #732
Merged
Conversation
This file contains 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
stephenswat
added
refactor
beomki-yeo
reviewed
Oct 9, 2024
beomki-yeo
approved these changes
Oct 14, 2024
stephenswat
force-pushed
the
refactor/ckf_liveness
branch
from
3264de6 to
2e5cf8b
Compare
|
Right now, we have two big kernels in the CKF. The first is `find_tracks`, which takes $N$ tracks and turns it into $M$ tracks, where $M$ can be either smaller than, equal to, or greater than $M$. The second kernel is the propagation kernel, which takes $N$ tracks and turns those into $M$ tracks, where $M$ is equal to or smaller than $N$, but never greater. The problem with kernels where the number of outputs differs from the number of inputs is that you need to atomically insert the elements into the output, and then you lose the ordering of your inputs. In this PR, I make propagation a 1:1 kernel, e.g. the size of the output is always equal to the size of the input. I do this by keeping track of which kernels are alive. This has the following benefits: 1. It makes the code easier to understand, as there is a bit less indirection. 2. It makes the code faster, although only by a small ~5% margin. 3. It promotes future optimization by preserving the order of tracks, i.e. we can perhaps try to do sorting only at the start. 4. This model will be useful for the proposed non-combinatorial KF.
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Outdated suggestions cannot be applied.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.
Right now, we have two big kernels in the CKF. The first is$N$ tracks and turns it into $M$ tracks, where $M$ can be either smaller than, equal to, or greater than $M$ . The second kernel is the propagation kernel, which takes $N$ tracks and turns those into $M$ tracks, where $M$ is equal to or smaller than $N$ , but never greater.
find_tracks, which takesThe problem with kernels where the number of outputs differs from the number of inputs is that you need to atomically insert the elements into the output, and then you lose the ordering of your inputs.
In this PR, I make propagation a 1:1 kernel, e.g. the size of the output is always equal to the size of the input. I do this by keeping track of which kernels are alive.
This has the following benefits: