FIX float counts in Paladin mapping#74
Merged
celiosantosjr merged 14 commits intomainfrom Jun 19, 2025
Merged
Conversation
- include a rounding in ngless counts to avoid split matches in Paladin
- Eliminate problems of compability
- do not test using Python 3.8 due to ONNX models
- eliminate ir10 model
- eliminate ir10 model
- Add downgraded IR versions of Hemo and AMP models to wide compatibility
- Downgrade of IR versions to 7 (allows compatibility to the present the installed environment and before ones)
- Include the same models adapted with:
```
import onnx
# Load the model
model = onnx.load("your_model.onnx")
# Downgrade IR version from 10 to 7
model.ir_version = 7
# Save the model
onnx.save(model, "downgraded_model.onnx")
```
- Downgrade version of the model to ONNX 16
- Dowgrade version to IR version 7
```
>>> from onnx import version_converter
>>> model = onnx.load('TEST.onnx')
>>> model.ir_version = 7
>>> model = optimize(model, ["eliminate_identity", "fuse_consecutive_transposes"])
>>> model = version_converter.convert_version(model, 16)
>>> onnx.save(model, "TEST.onnx")
```
Key Changes:
Added filter() step with samflag(secondary=False) which is the NGless equivalent of -F 256
The filtered alignments are then passed to count()
Maintained your existing integer conversion and output
This was
linked to
issues
Jun 19, 2025
celiosantosjr
added
bug
luispedro
added a commit
to luispedro/macrel
that referenced
this pull request
Nov 24, 2025
Big change is switching to ONNX for model storage to avoid depending on very particular versions of scikit-learn - Fix float counts in Paladin mapping (BigDataBiology#74) - Fix some issues in MacOS
luispedro
added a commit
to luispedro/macrel
that referenced
this pull request
Nov 24, 2025
Big change is switching to ONNX for model storage to avoid depending on very particular versions of scikit-learn - Fix float counts in Paladin mapping (BigDataBiology#74) - Fix some issues in MacOS
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
1 participant
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.
Paladin mapping does some frame-shitfs and probabilistic alignment, which leads to fractional weights propagating into SAM/BAM files. To fix that:
sam filtering to avoid this problem keeping only uniquely mapped matchers
count should consider only uniquely mapped matchers.
dowgraded version and IR version of onnx models to wider compatibility