ube4b · Sequence CNN

reference_ube4b_cnn_modulof0_20261104-144658

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	2.6037	0.6063	0.5696	0.3586	5520.0000	0.4904	26.0000	0.9000	3.6763	6.1880	0.5933	0.9400	4.2125	8.6246	0.3000	0.7502	0.9600	4.0389	8.6246	0.3700	0.7775	0.9347	3.6382	8.6246	0.5020	0.8006
validation	3.1721	-0.0919	-0.0719	-0.0415	571.0000	-0.0534	1659.0000	0.1000	-2.6759	0.3481	0.0000	-0.2274	0.1000	-2.5575	1.1347	0.0000	0.0000	-0.2666	0.1100	-2.4262	1.1347	0.0000	0.0000	-0.2685	0.0968	-2.7946	0.4566	0.0000	0.0000	-0.2940
test	3.4261	0.0223	0.0470	-0.0499	645.0000	0.1392	1687.0000	0.3000	-0.9119	1.8351	0.0000	0.0555	0.2400	-1.3381	1.8351	0.0000	0.0000	-0.0243	0.2000	-1.4369	2.7046	0.0200	0.0000	-0.0526	0.2500	-1.1222	1.8351	0.0000	0.0000	0.0213

Loss per epoch

training validation

Config

name: reference_ube4b_cnn_modulof0
task: dms
data:
  dataset: ube4b
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: modulo
  split_fold: 0
model:
  name: cnn
  conv:
  - - 128
    - 3
  - - 128
    - 3
  - - 128
    - 3
  - - 128
    - 3
  - - 128
    - 3
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7