update name
9aa418f
Running
MedGRPO Team
commited on
Commit History
update
0562da7
MedGRPO Team
commited on
improve the prompt
758710c
MedGRPO Team
commited on
update
4752404
MedGRPO Team
commited on
upload prediction only
b28cd8f
MedGRPO Team
commited on
Fix evaluate_all_pai to pass --skip-llm-judge to task main() functions
18339c0
MedGRPO Team
commited on
Fix eval_dvc.py main() to support --skip-llm-judge flag
5f41159
MedGRPO Team
commited on
Fix DVC evaluation to compute temporal F1 when --skip-llm-judge is set
dd1b9c6
MedGRPO Team
Claude Sonnet 4.5
commited on
Add semantic similarity matching for Next Action evaluation
a66b9a4
MedGRPO Team
Claude Sonnet 4.5
commited on
Fix TAL overall metrics computation and extraction
c8f4cad
MedGRPO Team
commited on
Use STG overall metrics instead of placeholder
d1337a3
MedGRPO Team
commited on
Fix STG evaluation to extract bbox_dict from struc_info
0b29eca
MedGRPO Team
commited on
Add placeholder for STG metrics when evaluation returns empty
6d00cf8
MedGRPO Team
commited on
Fix STG and next_action metric printing in overall mode
049c07c
MedGRPO Team
commited on
Show evaluation metrics even in silent mode (--grouping overall)
2323031
MedGRPO Team
commited on
Filter DVC/VS/RC from tasks list when skip-llm-judge is set
2bd924c
MedGRPO Team
commited on
Remove debug messages - system working correctly
80e3e7d
MedGRPO Team
commited on
Add debug inside loop to verify entry
d53d0f7
MedGRPO Team
commited on
Add debug at line 741
d878ea1
MedGRPO Team
commited on
Add debug before task list print
c886e23
MedGRPO Team
commited on
Debug before/after imports
4e448a4
MedGRPO Team
commited on
Add debug at function entry and after analyze
c736ac8
MedGRPO Team
commited on
Add explicit flush debug
1c45d6c
MedGRPO Team
commited on
update
f0846a5
MedGRPO Team
commited on
update
1af117c
MedGRPO Team
commited on
update
a605ebb
MedGRPO Team
commited on
Implement secure ground truth with prediction-only submission format
fe743f5
MedGRPO Team
commited on
Add support for pre-computed LLM judge scores
31817d3
MedGRPO Team
Claude Sonnet 4.5
commited on
Update evaluation metrics and leaderboard display
a36b7fe
MedGRPO Team
Claude Sonnet 4.5
commited on
Complete evaluator fixes for all 8 tasks
331979f
MedGRPO Team
commited on
Fix syntax errors and add TAL wrapper functions
8c805bc
MedGRPO Team
commited on
Fix evaluate_all_pai.py to use eval_caption_llm_judge
58dd6d7
MedGRPO Team
commited on
Consolidate evaluation scripts and remove hardcoded paths
3ea8a3a
MedGRPO Team
commited on
Remove all captioning_metrics dependencies
5a5d9ce
MedGRPO Team
commited on
Remove unused captioning_metrics folder
a09374a
MedGRPO Team
commited on
Add server-side LLM judge for caption evaluation
da2e674
MedGRPO Team
commited on
Make evaluation scripts fully self-contained
82f81ab
MedGRPO Team
commited on
Optimize evaluation scripts - remove optional files
f986294
MedGRPO Team
Claude Sonnet 4.5
commited on
Copy evaluation scripts to leaderboard and clean up template code
ba8d0d4
MedGRPO Team
Claude Sonnet 4.5
commited on