bgl3 · Fully connected

reference_bgl3_fc_plm_contiguousf1_20261104-181614

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	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	1.8545	0.3157	0.2664	0.0520	5657.0000	-0.0688	3980.0000	0.8000	0.7593	2.2054	0.3186	0.6000	0.1672	2.2054	0.0000	0.1685	0.5400	0.0787	2.2054	0.0000	0.1488	0.5267	0.0745	2.5002	0.0133	0.1530
validation	2.0600	0.1933	0.1136	0.0352	1465.0000	-0.0390	1591.0000	0.2000	-0.1666	2.1466	0.1010	0.4200	-0.3763	2.1466	0.0200	0.0413	0.3700	-0.4591	2.3136	0.0300	0.0132	0.4884	-0.1410	2.1466	0.0000	0.1217
test	1.7664	0.2917	0.2914	-0.0664	1203.0000	0.0039	1852.0000	0.6000	0.5009	2.0174	0.2986	0.4800	0.1847	2.3136	0.0400	0.2008	0.3700	0.0368	2.3136	0.0300	0.1526	0.4706	0.2264	2.0174	0.0000	0.2108

Loss per epoch

training validation

Config

name: reference_bgl3_fc_plm_contiguousf1
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: contiguous
  split_fold: 1
model:
  name: fcn
  hidden:
  - 100
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 32
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7