bgl3 · Linear

reference_bgl3_lr_mr5plm_randomf2_20261104-222627

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	0.5135	0.8553	0.8691	0.7242	124.0000	0.0599	2.0000	0.6000	0.0466	0.9971	0.1000	0.7327	0.3000	-0.1016	0.9971	0.1200	0.7774	0.2500	-0.1244	0.9971	0.2000	0.8195	0.3824	-0.0531	0.9971	0.1176	0.7708
validation	0.9934	0.6887	0.7176	0.4712	40.0000	0.0438	169.0000	0.1000	-0.4161	0.0601	0.0000	0.6258	0.1600	-0.3123	0.3302	0.1800	0.7629	0.1400	-0.3712	0.3302	0.3500	0.7734	0.0909	-0.3896	0.0601	0.0000	0.6423
test	0.9746	0.7001	0.7282	0.4843	53.0000	-0.0887	142.0000	0.1000	-0.7954	0.0950	0.0000	0.3977	0.2200	-0.4865	0.2821	0.2000	0.6386	0.1900	-0.4094	0.5199	0.3200	0.7258	0.0909	-0.7307	0.0950	0.0000	0.4316

Loss per epoch

training validation

Config

name: reference_bgl3_lr_mr5plm_randomf2
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: random
  split_fold: 2
  min_reads: 5
model:
  name: lr
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7