pab1 · Fully connected

reference_pab1_fc_contiguousf1_20261104-154420

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	0.9250	0.8713	0.8487	0.6788	2849.0000	0.0998	13323.0000	0.8000	0.3077	0.6696	0.6604	0.7000	0.1550	0.8601	0.0200	0.6703	0.6800	0.0849	0.9715	0.0400	0.6744	0.6768	0.0803	0.9715	0.0606	0.7118
validation	4.5126	-0.2196	-0.2104	-0.0624	475.0000	0.0537	4503.0000	0.0000	-4.6671	-2.9441	0.0000	-0.8207	0.0000	-4.5430	-0.2444	0.0000	0.0000	-0.8365	0.0200	-4.0186	0.4534	0.0100	0.0000	-0.6669	0.0000	-4.6884	-0.2444	0.0000	0.0000	-0.8881
test	3.6517	-0.0353	-0.0153	-0.0174	196.0000	0.2074	945.0000	0.0000	-3.6490	-0.8112	0.0000	-0.6322	0.0800	-2.0538	0.3028	0.0400	0.0000	-0.0050	0.0900	-1.8963	0.5233	0.0600	0.0000	0.0641	0.0417	-2.3256	0.0155	0.0000	0.0000	-0.1162

Loss per epoch

training validation

Config

name: reference_pab1_fc_contiguousf1
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: contiguous
  split_fold: 1
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