Matthew Ford
chore: v11.6 + v11.5 + candidate-20260716a artifact reports and manifests
371c1ae
Raw
History Blame Contribute Delete
6.13 kB
{
"seed": 2026,
"n_folds": 5,
"tasks": {
"odor_threshold": {
"n": 528,
"representations": {
"morgan": {
"spearman_per_fold": [
0.5891664895239047,
0.5761181722035544,
0.49085465889747004,
0.4690533355008412,
0.5160615744495983
],
"spearman_mean": 0.5282508461150737,
"mae_per_fold": [
0.920720612916806,
0.9434428660415641,
0.9756876260760686,
0.946392850057442,
0.9973405656429867
],
"mae_mean": 0.9567169041469736
},
"real_pom": {
"spearman_per_fold": [
0.5899730135314767,
0.692431053320434,
0.6125925941206682,
0.5154030327533157,
0.5966725252375984
],
"spearman_mean": 0.6014144437926985,
"mae_per_fold": [
0.8541821281132047,
0.9083183489566975,
0.8144851870660104,
0.8783027581554694,
0.9295570826481118
],
"mae_mean": 0.8769691009878986
},
"pom_plus_physics": {
"spearman_per_fold": [
0.6181704334523194,
0.6773959786255335,
0.6167143183201412,
0.5311499718319211,
0.6170832396888719
],
"spearman_mean": 0.6121027883837574,
"mae_per_fold": [
0.8374607211559753,
0.8909581452346482,
0.8159470532422318,
0.8634553789443405,
0.9063195930608696
],
"mae_mean": 0.8628281783276129
}
},
"paired_deltas": {
"pom_plus_physics_vs_morgan": {
"spearman_mean_delta": 0.08385194226868367,
"ci95": [
0.054993625507799015,
0.11105931938571481
]
},
"real_pom_vs_morgan": {
"spearman_mean_delta": 0.07316359767762484,
"ci95": [
0.03301643007841398,
0.11134251669363111
]
}
}
},
"substantivity": {
"n": 213,
"representations": {
"morgan": {
"spearman_per_fold": [
0.4245057707616193,
0.33147819849420457,
0.43029557351767717,
0.37044582064857523,
0.4051364206287573
],
"spearman_mean": 0.39237235681016674,
"mae_per_fold": [
0.47827800128869014,
0.5434274016539017,
0.5174299986940224,
0.5812849044186478,
0.5673127592488246
],
"mae_mean": 0.5375466130608173
},
"real_pom": {
"spearman_per_fold": [
0.5964737702977657,
0.7413485625205778,
0.6526065879951563,
0.7599812488520136,
0.6455531076418417
],
"spearman_mean": 0.679192655461471,
"mae_per_fold": [
0.3609498425656071,
0.3798304299890093,
0.3944361714313578,
0.36865210872025356,
0.43228461528503664
],
"mae_mean": 0.3872306335982528
},
"pom_plus_physics": {
"spearman_per_fold": [
0.5751197518613089,
0.8001616586361934,
0.6901056909449073,
0.7715846354189406,
0.6335803210126024
],
"spearman_mean": 0.6941104115747905,
"mae_per_fold": [
0.5776780778068328,
0.36386590247223566,
0.38406079745378496,
0.360649020125982,
0.43813927436367045
],
"mae_mean": 0.42487861444450115
}
},
"paired_deltas": {
"pom_plus_physics_vs_morgan": {
"spearman_mean_delta": 0.3017380547646238,
"ci95": [
0.2035851760788599,
0.4000203992224656
]
},
"real_pom_vs_morgan": {
"spearman_mean_delta": 0.2868202986513043,
"ci95": [
0.20579494300806705,
0.3678456542945415
]
}
}
},
"descriptor": {
"n": 339,
"n_classes": 25,
"representations": {
"morgan": {
"macro_f1_per_fold": [
0.10442230183609492,
0.12353136629452419,
0.15895316804407714,
0.13842623179007618,
0.1573417671162032
],
"macro_f1_mean": 0.1365349670161951,
"auroc_per_fold": []
},
"real_pom": {
"macro_f1_per_fold": [
0.29409937888198756,
0.1974747474747475,
0.38538348672802447,
0.20726090304130232,
0.36976087781041345
],
"macro_f1_mean": 0.29079587878729507,
"auroc_per_fold": []
},
"pom_plus_physics": {
"macro_f1_per_fold": [
0.25709499321000023,
0.20760324331752902,
0.3371748699479792,
0.20736239217100458,
0.37273762406447064
],
"macro_f1_mean": 0.27639462454219677,
"auroc_per_fold": []
}
},
"paired_deltas": {
"pom_plus_physics_vs_morgan": {
"macro_f1_mean_delta": 0.13985965752600163,
"ci95": [
0.09173775323635436,
0.18798156181564885
]
},
"real_pom_vs_morgan": {
"macro_f1_mean_delta": 0.15426091177109993,
"ci95": [
0.09504663638175831,
0.21347518716044153
]
}
}
},
"substitution": {
"n_pairs": 71,
"representations": {
"morgan": {
"median_rank": 4.0,
"top5_accuracy": 0.8
},
"real_pom": {
"median_rank": 4.0,
"top5_accuracy": 1.0
},
"pom_plus_physics": {
"median_rank": 5.0,
"top5_accuracy": 0.6181818181818182
}
}
}
}
}