pab1 · Sequence CNN

reference_pab1_cnn_plm_modulof2_20261104-184206

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	0.6277	0.9184	0.8963	0.8343	1717.0000	0.0298	4427.0000	0.3000	-0.0632	0.2206	0.6158	0.5400	-0.0135	0.4210	0.0000	0.6925	0.5200	-0.0426	0.8532	0.0100	0.7203	0.5294	-0.0146	0.8532	0.0261	0.7557
validation	1.3789	0.7782	0.7414	0.6055	223.0000	0.1092	529.0000	0.3000	-0.2043	0.3577	0.6140	0.3200	-0.2415	0.6443	0.1000	0.6757	0.3300	-0.2270	0.7491	0.1800	0.7221	0.3750	-0.1009	0.6443	0.0625	0.6813
test	1.8641	0.7319	0.7477	0.4807	116.0000	-0.2417	406.0000	0.0000	-0.5177	-0.1004	0.5772	0.1600	-0.3791	0.1479	0.0000	0.6969	0.1800	-0.4111	0.4214	0.1300	0.7181	0.1429	-0.3780	0.1479	0.0000	0.6458

Loss per epoch

training validation

Config

name: reference_pab1_cnn_plm_modulof2
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: modulo
  split_fold: 2
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