pab1 · Fully connected

reference_pab1_fc_plm_contiguousf0_20261104-175737

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	0.3838	0.9431	0.9047	0.8819	3428.0000	0.0616	1476.0000	0.9000	0.2153	0.5166	0.6370	0.8200	0.1953	0.8601	0.0200	0.6961	0.7800	0.1708	0.8698	0.0500	0.7191	0.7143	0.1403	1.0896	0.0779	0.7541
validation	1.7966	0.7147	0.6894	0.4995	196.0000	0.0410	1571.0000	0.1000	-0.3095	0.0916	0.6815	0.1400	-0.4231	0.2232	0.0200	0.6879	0.2200	-0.4405	0.3224	0.0900	0.7047	0.1250	-0.3811	0.2232	0.0000	0.6803
test	1.9833	0.7258	0.7287	0.4649	222.0000	-0.0266	2038.0000	0.2000	-0.2535	0.0634	0.7620	0.2000	-0.2862	0.2373	0.0200	0.7881	0.1800	-0.3699	0.3723	0.0600	0.7842	0.2041	-0.2856	0.2373	0.0204	0.7876

Loss per epoch

training validation

Config

name: reference_pab1_fc_plm_contiguousf0
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: contiguous
  split_fold: 0
model:
  name: fcn
  hidden:
  - 100
  - 100
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.001
  patience: 15
  min_delta: 1.0e-05
  seed: 7