bgl3 · Fully connected

reference_bgl3_fc_mr5plm_randomf0_20261104-222216

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	0.5811	0.8375	0.8468	0.6906	134.0000	0.0683	72.0000	0.2000	-0.1094	0.1391	0.6710	0.3600	-0.0799	0.7042	0.1400	0.7985	0.3300	-0.0970	1.6908	0.2500	0.8407	0.4412	-0.0132	0.7042	0.1176	0.8015
validation	0.8259	0.7557	0.7686	0.5654	37.0000	0.1780	118.0000	0.0000	-0.3433	-0.0549	0.6376	0.1400	-0.3363	0.4161	0.1600	0.7428	0.1300	-0.3530	0.4161	0.3300	0.7836	0.0909	-0.2991	0.1421	0.0000	0.6620
test	0.8593	0.7286	0.7418	0.5268	46.0000	0.0033	335.0000	0.1000	-0.6555	0.1248	0.4428	0.1400	-0.3909	0.1579	0.1800	0.6810	0.1300	-0.4621	0.7630	0.2500	0.6880	0.0909	-0.6179	0.1248	0.0000	0.4662

Loss per epoch

training validation

Config

name: reference_bgl3_fc_mr5plm_randomf0
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: random
  split_fold: 0
  min_reads: 5
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