ube4b · Graph CNN

reference_ube4b_gcn_modulof0_20261104-144750

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	4.0339	0.2575	0.1825	0.0063	5520.0000	0.3015	719.0000	0.6000	1.2049	4.0615	0.3009	0.7800	1.5880	6.5047	0.0600	0.3955	0.7300	1.4232	6.5047	0.0500	0.3993	0.7347	1.3482	6.5047	0.1143	0.4322
validation	3.0646	-0.0524	-0.0581	-0.0062	571.0000	0.0576	592.0000	0.0000	-2.0233	-0.2539	0.0000	-0.1034	0.1600	-1.4784	3.8996	0.0200	0.0000	0.0002	0.1800	-1.4519	4.3952	0.0500	0.0000	0.0078	0.1613	-1.5413	1.6266	0.0000	0.0000	-0.0138
test	3.2831	-0.0432	-0.0449	-0.0061	645.0000	0.0116	676.0000	0.0000	-2.3407	-1.8653	0.0000	-0.1954	0.1000	-2.2196	0.8717	0.0000	0.0000	-0.2181	0.1400	-1.8994	3.0492	0.0400	0.0000	-0.1690	0.1071	-2.2952	0.8717	0.0000	0.0000	-0.2148

Loss per epoch

training validation

Config

name: reference_ube4b_gcn_modulof0
task: dms
data:
  dataset: ube4b
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 7
  split_scheme: modulo
  split_fold: 0
model:
  name: gcn
  filters:
  - 128
  - 128
  - 128
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7