rpotter6298
280060db82
- Introduced multiple regression experiment configurations targeting vf_md, including: - cd_solo_reg_set.json: CD tower only regression setup. - img_solo_reg_set.json: Image tower only regression setup. - reg_head_epoch_sweep.json: Baseline regression sweeps at different epochs (50, 75, 100). - reg_head_set.json: Various regression setups including baseline and OrthoBridge configurations. - single_eye_reg.json: Single-eye regression setup for worst-eye aggregation analysis. - Added ensemble configurations for OrthoBridge with different inner bridges: - ortho_alts_ensemble.json: Ensemble tests with ConcatBridge, PairwiseAdditiveBridge, and GatedAdditiveBridge. - ortho_alts_tritower.json: Tritower tests with the same inner bridges. - Created V2-M specific configurations: - baseline_reg_nt50.json: Regression baseline with V2-M backbone. - geom_vec_gt.json and geom_vec_unet.json: Geometry vector injection experiments with V2-M. - single_l1_bridges.json: Single-eye ensemble experiments with various bridge types. - tritower_geom_gt.json: Tritower setup with GT contour-rasterized masks. - Promoted existing experiments to higher repetitions for robustness.
8 lines
507 B
JSON
8 lines
507 B
JSON
[
|
|
{
|
|
"_note": "Geometry-vector injection from UNet-derived segmentation, bumped to 10 reps to round out the refugelike geometry panel alongside ensemble_fused/no_geom (0.8979, n=10), ensemble_fused/geom_gt (0.9121, n=10), tri_v1/baseline_tri (0.8932, n=10), and tri_v1/baseline_solo (0.6833, n=10). Use the existing ensemble_fused_geom_unet.json base config which already wires up the EPC geometry_vectors channel.",
|
|
"run_name": "experiments/phase6_v4/ensemble_geom_vec_unet",
|
|
"reps": 10
|
|
}
|
|
]
|