ophthalmic_retinal · fundus_imaging

Fundus Artery Average Width

Valid comparisons for this clinically grounded task are grouped by fixed information set.

Per-track results

Benchmark Tracks

fundus_artery_average_width/00ada4c70237

Scored split: test. Primary metric: r2 (higher is better).

#	Predictor	Method	Absolute test R²	Test n
1	HPP Curated Panel	TabPFN 3.5	0.1059	1038
2	HPP Curated Panel	TabSwift	0.1058	1038
3	HPP Curated Panel	TabFM	0.1057	1038
4	HPP Curated Panel	TabPFN v2	0.1048	1038
5	HPP Curated Panel	TabICL	0.1041	1038
6	HPP Curated Panel	TabDPT	0.1029	1038
7	HPP Curated Panel	RealMLP	0.1022	1038
8	HPP Curated Panel	Gradient-boosted trees	0.1016	1038
9	HPP Curated Panel	Ridge	0.0996	1038

fundus_artery_average_width/1ff49b99ee43

Scored split: test. Primary metric: r2 (higher is better).

#	Predictor	Method	Absolute test R²	Test n
1	CGM Iglu	RealMLP	0.09504	928
2	CGM Iglu	TabPFN 3.5	0.09435	928
3	CGM Iglu	Ridge	0.09323	928
4	CGM Iglu	TabICL	0.09272	928
5	CGM Iglu	Gradient-boosted trees	0.09235	928
6	CGM Iglu	TabDPT	0.09161	928
7	CGM Iglu	TabPFN v2	0.08984	928
8	CGM Iglu	TabFM	0.08888	928
9	CGM Iglu	TabSwift	0.08555	928
10	Participant Embedding	Ridge	0.07246	928

fundus_artery_average_width/2140c729851e

Scored split: test. Primary metric: r2 (higher is better).

#	Predictor	Method	Absolute test R²	Test n
1	Participant Embedding	Ridge	0.06469	564

fundus_artery_average_width/617c9e1db03f

Scored split: test. Primary metric: r2 (higher is better).

#	Predictor	Method	Absolute test R²	Test n
1	HPP Curated Panel	TabDPT	0.09544	463
2	HPP Curated Panel	RealMLP	0.09421	463
3	HPP Curated Panel	TabPFN 3.5	0.09404	463
4	HPP Curated Panel	TabPFN v2	0.09386	463
5	HPP Curated Panel	TabICL	0.09166	463
6	HPP Curated Panel	Ridge	0.09129	463
7	HPP Curated Panel	Gradient-boosted trees	0.09116	463
8	HPP Curated Panel	TabFM	0.09088	463
9	HPP Curated Panel	TabSwift	0.087	463

fundus_artery_average_width/974fedf9cd0d

Scored split: test. Primary metric: r2 (higher is better).

#	Predictor	Method	Absolute test R²	Test n
1	HPP Curated Panel	TabDPT	0.1201	804
2	HPP Curated Panel	Ridge	0.1183	804
3	HPP Curated Panel	TabPFN 3.5	0.118	804
4	HPP Curated Panel	TabPFN v2	0.1172	804
5	HPP Curated Panel	TabICL	0.1146	804
6	HPP Curated Panel	TabFM	0.1141	804
7	HPP Curated Panel	Gradient-boosted trees	0.1124	804
8	HPP Curated Panel	RealMLP	0.1091	804
9	HPP Curated Panel	TabSwift	0.1066	804

fundus_artery_average_width/e1495841a055

Scored split: test. Primary metric: r2 (higher is better).

#	Predictor	Method	Absolute test R²	Test n
1	HPP Curated Panel	TabSwift	0.1137	552
2	HPP Curated Panel	Ridge	0.1125	552
3	HPP Curated Panel	TabPFN 3.5	0.1113	552
4	HPP Curated Panel	TabICL	0.1094	552
5	HPP Curated Panel	TabDPT	0.109	552
6	HPP Curated Panel	TabFM	0.1085	552
7	HPP Curated Panel	RealMLP	0.1073	552
8	HPP Curated Panel	Gradient-boosted trees	0.1051	552
9	HPP Curated Panel	TabPFN v2	0.1039	552

fundus_artery_average_width/e4537dafb19b

Scored split: test. Primary metric: r2 (higher is better).

#	Predictor	Method	Absolute test R²	Test n
1	HPP Curated Panel	TabSwift	0.1204	965
2	HPP Curated Panel	TabDPT	0.1204	965
3	HPP Curated Panel	TabPFN 3.5	0.12	965
4	HPP Curated Panel	TabICL	0.1189	965
5	HPP Curated Panel	TabPFN v2	0.1187	965
6	HPP Curated Panel	TabFM	0.118	965
7	HPP Curated Panel	RealMLP	0.1157	965
8	HPP Curated Panel	Gradient-boosted trees	0.1132	965
9	HPP Curated Panel	Ridge	0.1071	965

fundus_artery_average_width/fe3d1761eadf

Scored split: test. Primary metric: r2 (higher is better).

#	Predictor	Method	Absolute test R²	Test n
1	Participant Embedding	Ridge	0.06998	610