bgl3 · Fully connected

reference_bgl3_fc_mr5_randomf1_20261104-221809

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	0.6963	0.7921	0.7969	0.6246	127.0000	0.2014	1.0000	0.4000	0.0328	0.9971	0.1000	0.7309	0.2400	-0.1779	0.9971	0.1000	0.7476	0.1800	-0.2588	0.9971	0.1700	0.7526	0.2647	-0.1457	0.9971	0.0882	0.7364
validation	1.1238	0.6397	0.6451	0.4054	53.0000	-0.0564	267.0000	0.1000	-0.8042	0.1135	0.0000	0.3938	0.1400	-0.7610	0.3940	0.1400	0.4919	0.1100	-0.8277	0.5199	0.1900	0.4870	0.0909	-0.7498	0.1135	0.0000	0.4229
test	1.1101	0.6490	0.6450	0.4158	37.0000	0.0868	702.0000	0.1000	-1.0698	0.3347	0.1000	0.2944	0.1200	-0.9356	0.4008	0.1200	0.4147	0.0900	-0.8502	0.4856	0.2000	0.4928	0.0909	-1.1316	0.3347	0.0909	0.2666

Loss per epoch

training validation

Config

name: reference_bgl3_fc_mr5_randomf1
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: random
  split_fold: 1
  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