You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Related to issue #86 domain_specific_metric_calculations
One application of domain-specific metric calculations is to only compare metrics for those domains for which we are confident that the anatomical/manual labels and transcriptomic signatures align. To do that, we need to specify the confidence of each cell's label and it's suitability for use in domain-specific metrics.
We need to add an optional, numerical column, label_confidence, to labels.tsv. For now, it should only be used as a boolean, 0 for "too low confidence, don't use this label for metrics" and 1 for "high enough confidence to use this label for metrics". In the future, if we want to expand to numeric rankings between 0 and 1, we will need to define explicit criteria for each numeric ranking.
TODO:
Add the optional numeric label_confidence column to labels.tsv templates/documentation.
The text was updated successfully, but these errors were encountered:
Related to issue #86 domain_specific_metric_calculations
One application of domain-specific metric calculations is to only compare metrics for those domains for which we are confident that the anatomical/manual labels and transcriptomic signatures align. To do that, we need to specify the confidence of each cell's label and it's suitability for use in domain-specific metrics.
We need to add an optional, numerical column,
label_confidence
, to labels.tsv. For now, it should only be used as a boolean, 0 for "too low confidence, don't use this label for metrics" and 1 for "high enough confidence to use this label for metrics". In the future, if we want to expand to numeric rankings between 0 and 1, we will need to define explicit criteria for each numeric ranking.TODO:
Add the optional numeric
label_confidence
column to labels.tsv templates/documentation.The text was updated successfully, but these errors were encountered: