pab1 · Sequence CNN

reference_pab1_cnn_modulof3_20261104-161448

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	3.0212	0.5075	0.5411	0.1800	1256.0000	0.0648	4121.0000	0.2000	-0.2028	0.2293	0.5956	0.1400	-0.5343	0.4542	0.0000	0.5364	0.1200	-0.5340	0.4542	0.0200	0.5684	0.1689	-0.4646	0.4681	0.0338	0.6125
validation	3.9579	-0.0495	-0.0186	-0.0646	234.0000	0.0190	902.0000	0.0000	-3.0278	-0.5477	0.0000	-0.6098	0.0600	-2.6526	0.2124	0.0000	0.0000	-0.4866	0.0800	-2.5099	0.2137	0.0400	0.0000	-0.4378	0.0667	-2.9388	0.0857	0.0000	0.0000	-0.5831
test	3.6756	0.2125	0.1778	-0.0516	223.0000	0.0044	931.0000	0.0000	-1.0853	-0.0295	0.0000	0.2456	0.1400	-0.9752	0.4974	0.0600	0.0000	0.3292	0.1100	-1.0687	0.4974	0.0600	0.0000	0.3016	0.0000	-1.4058	-0.0295	0.0000	0.0000	0.1156

Loss per epoch

training validation

Config

name: reference_pab1_cnn_modulof3
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: modulo
  split_fold: 3
model:
  name: cnn
  conv:
  - - 128
    - 17
  - - 128
    - 17
  - - 128
    - 17
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7