New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
DestVI changed l1 loss. Expected proportions #1591
base: main
Are you sure you want to change the base?
Conversation
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## main #1591 +/- ##
==========================================
- Coverage 89.18% 88.94% -0.25%
==========================================
Files 152 152
Lines 12482 12536 +54
==========================================
+ Hits 11132 11150 +18
- Misses 1350 1386 +36 ☔ View full report in Codecov by Sentry. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Optional Obsm field implementation looks good to me
@cane11 any reason to keep this open? |
365a4a0
to
34e03d6
Compare
@martinkim0 It's the version of destVI we are using for Curio data (other sparsity regularization). Can you have a pass? |
for more information, see https://pre-commit.ci
…-tools into can-destvi-expected
@@ -78,7 +78,7 @@ def __init__( | |||
n_latent: int, | |||
n_layers: int, | |||
dropout_decoder: float, | |||
l1_reg: float, | |||
celltype_reg: dict, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I feel like this argument would be clearer if we separate it out into two arguments: celltype_regularization: Literal["l1", "entropy"] | None = None
and celltype_regularization_weight: float | None = None
Uses proportions computed in other deconvolution method to infer cell activation state of those cells. Application activation for e.g. Cell2Location results. L1 loss changed. Still in discussion.
Optional obsm field.