bgl3 · Fully connected

reference_bgl3_fc_plm_modulof0_20261104-175032

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	1.6655	0.3000	0.2572	0.0736	3841.0000	-0.0568	2585.0000	0.8000	0.6500	1.4833	0.3087	0.6000	0.2259	2.0841	0.0000	0.2028	0.5500	0.1356	2.3136	0.0200	0.1819	0.5364	0.1273	2.3637	0.0273	0.1804
validation	1.3787	0.3622	0.3103	0.1193	424.0000	-0.1081	577.0000	0.5000	0.3869	1.3581	0.2844	0.4000	0.0805	2.3136	0.0600	0.2097	0.4300	0.0905	2.4113	0.1100	0.2405	0.5333	0.4269	1.8690	0.0000	0.3077
test	1.3623	0.3098	0.3000	0.0956	338.0000	-0.2463	1040.0000	0.4000	-0.8894	0.2173	-0.1182	0.2200	-0.6433	1.9459	0.0200	-0.0330	0.3000	-0.4027	1.9459	0.0500	0.0766	0.3077	-0.7809	0.2173	0.0000	-0.0800

Loss per epoch

training validation

Config

name: reference_bgl3_fc_plm_modulof0
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: modulo
  split_fold: 0
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