ube4b · Sequence CNN

reference_ube4b_cnn_modulof2_20261104-163703

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	3.1800	0.3520	0.3698	0.0387	5464.0000	0.1193	5235.0000	0.6000	0.2272	2.3674	0.2179	0.5400	0.0676	2.6790	0.0000	0.2368	0.5000	-0.1265	3.3952	0.0000	0.2273	0.5233	-0.0664	4.6933	0.0394	0.2771
validation	3.0114	0.1041	0.1024	0.0057	418.0000	0.0687	1030.0000	0.4000	-0.7767	1.2563	0.1115	0.2600	-1.2140	3.3635	0.0400	0.0235	0.2900	-0.9837	3.3635	0.0700	0.1014	0.2778	-0.9987	1.7218	0.0000	0.0693
test	4.8183	0.0780	0.0919	-0.0019	819.0000	-0.0387	2340.0000	0.2000	-1.0963	0.4962	0.0048	0.2400	-0.9665	3.8909	0.0000	0.0268	0.2600	-1.0117	3.8909	0.0100	0.0215	0.2500	-1.0650	3.8909	0.0000	0.0102

Loss per epoch

training validation

Config

name: reference_ube4b_cnn_modulof2
task: dms
data:
  dataset: ube4b
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: modulo
  split_fold: 2
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