pab1 · Sequence CNN

reference_pab1_cnn_plm_contiguousf3_20261105-040830

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	0.9021	0.8903	0.8905	0.7726	1489.0000	0.0313	1889.0000	0.1000	-0.1748	0.0520	0.7045	0.4200	-0.0414	0.3747	0.0000	0.7916	0.4800	-0.0348	0.3747	0.0000	0.8160	0.4766	-0.0384	0.6046	0.0794	0.8403
validation	1.6757	0.7291	0.6155	0.1016	931.0000	0.0349	1367.0000	0.2000	-0.2098	0.2331	0.3911	0.2200	-0.3252	0.8601	0.0400	0.3858	0.2400	-0.3312	0.8601	0.0200	0.4129	0.2000	-0.3693	0.7491	0.0222	0.3569
test	1.3613	0.7850	0.6939	0.4636	383.0000	-0.0504	2014.0000	0.1000	-0.2907	0.0138	0.4514	0.2000	-0.2019	0.4637	0.0400	0.5698	0.2400	-0.1750	0.4637	0.0700	0.6338	0.1875	-0.2107	0.4333	0.0000	0.5414

Loss per epoch

training validation

Config

name: reference_pab1_cnn_plm_contiguousf3
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: contiguous
  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