ube4b · Linear

reference_ube4b_lr_plm_modulof1_20261104-181501

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	2.3678	0.5340	0.5691	0.2762	4756.0000	0.1963	738.0000	1.0000	3.8509	6.1279	0.2000	0.7106	0.9600	2.8057	6.1279	0.1000	0.6837	0.9300	2.3854	6.1279	0.1700	0.6740	0.8419	1.6843	6.3990	0.2692	0.6249
validation	4.6259	0.2827	0.2763	0.0381	819.0000	0.0096	2107.0000	0.4000	-0.4179	2.2738	0.0000	0.0996	0.3600	-0.2361	3.1584	0.0000	0.1458	0.3800	-0.2840	3.1584	0.0000	0.1529	0.3438	-0.1962	3.1584	0.0000	0.1450
test	2.8207	0.2922	0.2914	0.0739	571.0000	0.0229	2351.0000	0.4000	-0.0122	1.7130	0.0000	0.2788	0.3800	-0.2317	2.7945	0.0600	0.3085	0.3700	-0.5360	2.7945	0.0600	0.2675	0.3871	-0.1172	2.1196	0.0323	0.3045

Loss per epoch

training validation

Config

name: reference_ube4b_lr_plm_modulof1
task: dms
data:
  dataset: ube4b
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: modulo
  split_fold: 1
model:
  name: lr
train:
  epochs: 300
  batch_size: 64
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7