pab1 · Fully connected

reference_pab1_fc_modulof3_20261104-161505

Metrics

Set	mse	pearsonr	spearmanr	r2	above_wt_n	above_wt_spearmanr	budget_to_max	hit_rate_at_10	mean_at_10	max_at_10	precision_at_10	captured_at_10	hit_rate_at_50	mean_at_50	max_at_50	precision_at_50	captured_at_50	hit_rate_at_100	mean_at_100	max_at_100	precision_at_100	captured_at_100	hit_rate_at_1pct	mean_at_1pct	max_at_1pct	precision_at_1pct	captured_at_1pct
train	1.9080	0.7297	0.7102	0.4821	1256.0000	0.0540	2131.0000	0.5000	-0.1375	0.3125	0.0000	0.6168	0.2800	-0.1148	0.4681	0.0000	0.6859	0.2700	-0.1738	0.4681	0.0200	0.7043	0.2500	-0.1615	0.8116	0.0203	0.7304
validation	3.7631	-0.1068	-0.0618	-0.0122	234.0000	0.1062	389.0000	0.4000	-1.5159	0.4714	0.1000	0.0000	0.0627	0.2000	-2.3277	0.4714	0.0600	0.0000	-0.3279	0.1800	-2.0791	0.4714	0.1200	0.0000	-0.2167	0.3333	-1.7982	0.4714	0.0667	0.0000	-0.0643
test	3.5314	-0.0986	-0.1380	-0.0103	223.0000	0.0859	1288.0000	0.0000	-0.6650	-0.2174	0.0000	0.0000	0.4213	0.0200	-1.7637	0.0859	0.0000	0.0000	-0.0431	0.0400	-1.8785	0.4465	0.0100	0.0000	-0.1030	0.0000	-1.3672	-0.2174	0.0000	0.0000	0.1323

Loss per epoch

training validation

Config

name: reference_pab1_fc_modulof3
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: modulo
  split_fold: 3
model:
  name: fcn
  hidden:
  - 100
  - 100
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.001
  patience: 15
  min_delta: 1.0e-05
  seed: 7