bgl3 · Fully connected

reference_bgl3_fc_plm_modulof1_20261104-181137

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	1.7403	0.2738	0.2191	0.0252	3546.0000	-0.0744	2531.0000	0.7000	0.6981	2.2054	0.3125	0.4400	-0.0543	2.2054	0.0000	0.1022	0.4200	-0.1588	2.2054	0.0300	0.0711	0.4141	-0.1682	2.2054	0.0303	0.0676
validation	1.4205	0.2893	0.2985	0.0785	432.0000	-0.0654	647.0000	0.4000	-0.8504	1.3581	-0.1111	0.3600	-0.3468	2.0174	0.0400	0.0762	0.3600	-0.2813	2.3637	0.1000	0.1131	0.4118	-0.5424	1.5945	0.0000	-0.0031
test	1.3945	0.3587	0.3101	0.1092	424.0000	-0.1081	573.0000	0.5000	0.3869	1.3581	0.2844	0.3800	0.0672	2.3136	0.0600	0.2043	0.4300	0.0708	2.4113	0.1000	0.2315	0.5333	0.3968	1.8690	0.0000	0.2969

Loss per epoch

training validation

Config

name: reference_bgl3_fc_plm_modulof1
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: modulo
  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