pab1 · Sequence CNN

reference_pab1_cnn_plm_modulof1_20261104-181329

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	0.6039	0.9168	0.8671	0.8319	1937.0000	0.0301	4998.0000	0.5000	-0.0028	0.2987	0.6356	0.5600	0.0046	0.5640	0.0000	0.6985	0.4800	-0.0358	0.6443	0.0200	0.7135	0.4497	-0.0544	0.6443	0.0268	0.7247
validation	1.4205	0.7793	0.7770	0.6043	116.0000	-0.2508	342.0000	0.1000	-0.3012	0.0605	0.6657	0.2600	-0.3595	0.2777	0.0200	0.7057	0.2200	-0.3233	0.4214	0.1800	0.7595	0.1429	-0.3408	0.0605	0.0000	0.6613
test	1.8708	0.7367	0.7382	0.5313	132.0000	-0.2243	601.0000	0.3000	-0.1123	0.0644	0.7431	0.1800	-0.2724	0.1281	0.0000	0.7750	0.1800	-0.2643	0.3930	0.1200	0.8134	0.2353	-0.1357	0.0644	0.0000	0.7670

Loss per epoch

training validation

Config

name: reference_pab1_cnn_plm_modulof1
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: modulo
  split_fold: 1
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