pab1 · Graph CNN

reference_pab1_gcn_contiguousf0_20261104-145739

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	2.0879	0.6598	0.5868	0.3576	3428.0000	0.0691	10538.0000	0.2000	-0.0918	0.4346	0.5232	0.2400	-0.1728	0.5871	0.0000	0.5453	0.2900	-0.1763	0.5871	0.0000	0.5701	0.2468	-0.2062	0.8601	0.0260	0.5952
validation	3.6519	-0.0166	0.0079	-0.0174	196.0000	0.0565	90.0000	0.0000	-4.0527	-1.5912	0.0000	-0.7910	0.0000	-3.8139	-0.0079	0.0000	0.0000	-0.7530	0.0200	-3.1262	0.6164	0.0200	0.0000	-0.4771	0.0000	-4.0895	-0.1819	0.0000	0.0000	-0.8387
test	4.4000	0.0349	0.0309	-0.1871	222.0000	0.1271	468.0000	0.0000	-2.7315	-0.6637	0.0000	-0.0407	0.0000	-2.9041	-0.6637	0.0000	0.0000	-0.1014	0.0000	-2.7860	-0.0697	0.0000	0.0000	-0.0632	0.0000	-2.9063	-0.6637	0.0000	0.0000	-0.1020

Loss per epoch

training validation

Config

name: reference_pab1_gcn_contiguousf0
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 7
  split_scheme: contiguous
  split_fold: 0
model:
  name: gcn
  filters:
  - 32
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7