pab1 · Sequence CNN

reference_pab1_cnn_randomf0_20261104-142750

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	0.2819	0.9620	0.9407	0.9247	2767.0000	0.1182	1169.0000	1.0000	0.1673	0.3286	0.6841	0.8200	0.1956	0.8505	0.0400	0.7544	0.7700	0.1553	0.8744	0.0500	0.7743	0.6980	0.1025	0.8744	0.0857	0.8021
validation	0.4744	0.9346	0.9180	0.8732	898.0000	0.0908	2839.0000	0.7000	0.1573	0.4835	0.0000	0.7110	0.6800	0.0249	0.8419	0.0400	0.0000	0.7422	0.6700	0.0641	0.8419	0.0700	0.0000	0.7946	0.6829	0.0683	0.8419	0.0732	0.0000	0.7849
test	0.4984	0.9319	0.9172	0.8678	851.0000	0.0792	16.0000	0.6000	0.0606	0.3098	0.0000	0.7364	0.6600	0.0590	0.8698	0.0200	0.0000	0.7843	0.6100	0.0385	0.8698	0.0500	0.0000	0.8078	0.5976	0.0331	0.8698	0.0488	0.0000	0.7962

Loss per epoch

training validation

Config

name: reference_pab1_cnn_randomf0
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: random
  split_fold: 0
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