bgl3 · Linear

reference_bgl3_lr_modulof3_20261104-161030

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	1.2955	0.6092	0.5812	0.3377	3467.0000	0.1613	3691.0000	0.9000	1.2491	2.0841	0.5026	0.8600	1.1381	2.9405	0.0400	0.5340	0.8700	1.0888	2.9405	0.0600	0.5449	0.8788	1.1055	2.9405	0.0606	0.5504
validation	1.3904	-0.0350	-0.0363	-0.0271	437.0000	-0.0535	1036.0000	0.4000	-0.1888	1.8690	0.0000	0.0722	0.2600	-0.5063	2.0841	0.0400	0.0000	-0.0469	0.2900	-0.4978	2.0841	0.0900	0.0000	-0.0479	0.3125	-0.2246	1.8690	0.0000	0.0000	0.0627
test	1.3854	0.0495	0.0730	-0.0048	437.0000	-0.0234	576.0000	0.7000	0.1655	1.5945	0.0000	0.1878	0.3600	-0.2158	1.5945	0.0200	0.0000	0.0804	0.3400	-0.2351	2.3136	0.1100	0.0000	0.0822	0.4375	-0.1298	1.5945	0.0000	0.0000	0.0965

Loss per epoch

training validation

Config

name: reference_bgl3_lr_modulof3
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: modulo
  split_fold: 3
model:
  name: lr
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7