pab1 · Sequence CNN

reference_pab1_cnn_plm_modulof0_20261104-175120

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	0.9989	0.8549	0.8126	0.7302	1755.0000	0.0080	3907.0000	0.8000	0.1160	0.4534	0.6818	0.4800	-0.0695	0.4637	0.0000	0.6719	0.4200	-0.1096	0.4637	0.0000	0.6859	0.3913	-0.1215	0.4637	0.0072	0.6951
validation	1.8492	0.7344	0.7359	0.5368	132.0000	-0.2300	584.0000	0.3000	-0.1204	0.0644	0.7403	0.2000	-0.2412	0.2405	0.0400	0.7874	0.1900	-0.2576	0.3930	0.1300	0.8161	0.1765	-0.1862	0.0644	0.0000	0.7485
test	1.3295	0.7942	0.7856	0.5862	171.0000	0.0211	329.0000	0.1000	-0.2047	0.2232	0.6428	0.2600	-0.2376	0.9033	0.0600	0.7326	0.2700	-0.2143	0.9033	0.1900	0.7843	0.1176	-0.2754	0.2232	0.0000	0.6497

Loss per epoch

training validation

Config

name: reference_pab1_cnn_plm_modulof0
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: modulo
  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