Actions: Lightning-AI/torchmetrics
Actions
Showing runs from all workflows
31,230 workflow runs
31,230 workflow runs
DiceScore
yields 1.0 scores when using average="none"
for samples where class is not present
Greetings
#7478:
Issue #2850
labeled
by
Borda
ignore_index
255 is not valid for inputs with 6 classes
Greetings
#7477:
Issue #2849
labeled
by
Borda
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Greetings
#7476:
Issue #2852
labeled
by
Borda
ignore_index
255 is not valid for inputs with 6 classes
Slash Command Dispatch
#1369:
Issue comment #2849 (comment)
created
by
robmarkcole
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Slash Command Dispatch
#1368:
Issue comment #2852 (comment)
created
by
AdrianM0
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Greetings
#7474:
Issue #2852
edited
by
AdrianM0
ignore_index
255 is not valid for inputs with 6 classes
Slash Command Dispatch
#1367:
Issue comment #2849 (comment)
created
by
Borda
DiceScore(average="weighted")
produces scores bigger than 1.0
Greetings
#7473:
Issue #2851
assigned
by
Borda
DiceScore(average="weighted")
produces scores bigger than 1.0
Greetings
#7472:
Issue #2851
labeled
by
Borda
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Slash Command Dispatch
#1366:
Issue comment #2852 (comment)
created
by
Borda
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Greetings
#7471:
Issue #2852
edited
by
AdrianM0
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Greetings
#7470:
Issue #2852
edited
by
AdrianM0
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Greetings
#7469:
Issue #2852
edited
by
AdrianM0
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Greetings
#7468:
Issue #2852
edited
by
AdrianM0
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Greetings
#7467:
Issue #2852
opened
by
AdrianM0
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Greetings
#7466:
Issue #2852
labeled
by
AdrianM0
RetrievalRecall
and RetrievalMRR
seem to function differently for world_size > 1
and top_k=1
Greetings
#7465:
Issue #2852
labeled
by
AdrianM0
GeneralizedDiceScore
yields 0 scores when using per_class=True
for samples where class is not present
Greetings
#7464:
Issue #2846
edited
by
nkaenzig