cassandra-asl-tram2 / results.json
cassandra-anon's picture
Initial release: CASSANDRA cassandra-asl-tram2 (anonymous CCS 2026 artifact)
4e766eb verified
Raw
History Blame Contribute Delete
4.02 kB
{
"dataset": "tram2",
"n_npzs": 6,
"per_seed": [
{
"path": "../models/asl_recipe_tram2_orig_probs/seed_42.npz",
"seed": 42,
"tag": "orig",
"chose": "swa",
"tau_0.5": {
"doc_f1": 0.7292129677362525,
"macro_f1": 0.5504301683233094,
"n_scored_docs": 30
}
},
{
"path": "../models/asl_recipe_tram2_orig_probs/seed_123.npz",
"seed": 123,
"tag": "orig",
"chose": "ema",
"tau_0.5": {
"doc_f1": 0.7436287853187492,
"macro_f1": 0.5701923642028678,
"n_scored_docs": 30
}
},
{
"path": "../models/asl_recipe_tram2_orig_probs/seed_456.npz",
"seed": 456,
"tag": "orig",
"chose": "base",
"tau_0.5": {
"doc_f1": 0.7333488198828281,
"macro_f1": 0.5704794001767052,
"n_scored_docs": 30
}
},
{
"path": "../models/asl_recipe_tram2_repl_probs/seed_789.npz",
"seed": 789,
"tag": "repl",
"chose": "swa",
"tau_0.5": {
"doc_f1": 0.7310552364563793,
"macro_f1": 0.5695919594127242,
"n_scored_docs": 30
}
},
{
"path": "../models/asl_recipe_tram2_repl_probs/seed_2024.npz",
"seed": 2024,
"tag": "repl",
"chose": "swa",
"tau_0.5": {
"doc_f1": 0.703092004352701,
"macro_f1": 0.5492939798102902,
"n_scored_docs": 30
}
},
{
"path": "../models/asl_recipe_tram2_repl_probs/seed_3141.npz",
"seed": 3141,
"tag": "repl",
"chose": "ema",
"tau_0.5": {
"doc_f1": 0.6962908440035396,
"macro_f1": 0.5171692881057236,
"n_scored_docs": 30
}
}
],
"per_tag_ensembles": {
"orig": {
"seeds": [
42,
123,
456
],
"tau_0.5": {
"doc_f1": 0.7669756310205698,
"macro_f1": 0.5932483600169265,
"bootstrap_95ci": {
"mean": 0.7669756310205698,
"lo": 0.70433739643844,
"hi": 0.8226708653932212
}
},
"global_tau_on_dev": {
"tau": 0.5900000000000002,
"dev_macro_f1": 0.5857674167735052,
"doc_f1": 0.7661424403704719,
"macro_f1": 0.6035583297232558
},
"per_class_tau_on_dev": {
"doc_f1": 0.7436602705641955,
"macro_f1": 0.5912694986450675,
"n_fit": 25,
"n_fallback": 25
},
"n_scored_docs": 30,
"n_skipped_empty_docs": 1
},
"repl": {
"seeds": [
789,
2024,
3141
],
"tau_0.5": {
"doc_f1": 0.7583401131075023,
"macro_f1": 0.6064399082533692,
"bootstrap_95ci": {
"mean": 0.7583401131075023,
"lo": 0.6934627779894106,
"hi": 0.8187075956468564
}
},
"global_tau_on_dev": {
"tau": 0.5900000000000002,
"dev_macro_f1": 0.6059690063802943,
"doc_f1": 0.7692314145719508,
"macro_f1": 0.6049801238258686
},
"per_class_tau_on_dev": {
"doc_f1": 0.7377386395365326,
"macro_f1": 0.5928088623707426,
"n_fit": 25,
"n_fallback": 25
},
"n_scored_docs": 30,
"n_skipped_empty_docs": 1
}
},
"combined_ensemble": {
"n_seeds": 6,
"seeds": [
42,
123,
456,
789,
2024,
3141
],
"tau_0.5": {
"doc_f1": 0.7691482816151715,
"macro_f1": 0.6237927709733632,
"bootstrap_95ci": {
"mean": 0.7691482816151715,
"lo": 0.7077068861888169,
"hi": 0.8248226631751874
}
},
"global_tau_on_dev": {
"tau": 0.6300000000000001,
"dev_macro_f1": 0.5978876773717535,
"doc_f1": 0.7742609511428918,
"macro_f1": 0.6114242954700899
},
"per_class_tau_on_dev": {
"doc_f1": 0.7541317504397019,
"macro_f1": 0.6110780048728575,
"n_fit": 25,
"n_fallback": 25
},
"n_scored_docs": 30,
"n_skipped_empty_docs": 1
}
}