pab1 · Graph CNN

reference_pab1_gcn_plm_contiguousf4_20261105-043820

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	0.9579	0.8785	0.8705	0.7695	1652.0000	-0.0133	2471.0000	0.2000	-0.1892	0.0153	0.6493	0.2200	-0.1908	0.4333	0.0000	0.6985	0.2200	-0.1743	0.4974	0.0300	0.7295	0.2667	-0.1565	0.4974	0.0222	0.7586
validation	1.7695	0.7428	0.7442	0.5226	222.0000	-0.0231	2035.0000	0.2000	-0.2819	0.0634	0.7528	0.1800	-0.2745	0.2373	0.0200	0.7921	0.1700	-0.3625	0.3723	0.0600	0.7868	0.1837	-0.2766	0.2373	0.0204	0.7907
test	1.1173	0.7531	0.6241	0.4010	931.0000	0.0351	1350.0000	0.2000	-0.2704	0.2331	0.3612	0.2200	-0.3298	0.8601	0.0400	0.3832	0.2300	-0.3403	0.8601	0.0200	0.4073	0.2000	-0.3633	0.8601	0.0444	0.3602

Loss per epoch

training validation

Config

name: reference_pab1_gcn_plm_contiguousf4
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  graph_threshold: 7
  split_scheme: contiguous
  split_fold: 4
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