bgl3 · Graph CNN

reference_bgl3_gcn_mr5_randomf1_20261104-221821

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	0.5938	0.8297	0.8316	0.6798	127.0000	0.0766	92.0000	0.3000	-0.1114	0.3302	0.6686	0.2200	-0.1226	0.3538	0.1200	0.7755	0.2100	-0.1327	0.9971	0.1700	0.8204	0.2059	-0.1385	0.3302	0.0588	0.7399
validation	0.9631	0.7004	0.7248	0.4904	53.0000	-0.1744	14.0000	0.4000	-0.6603	0.2595	0.4584	0.2000	-0.3928	1.6908	0.2000	0.6888	0.1600	-0.4152	1.6908	0.2600	0.7224	0.4545	-0.5877	0.2595	0.0000	0.4965
test	0.9170	0.7195	0.7243	0.5174	37.0000	0.0484	500.0000	0.0000	-0.9168	-0.0470	0.3667	0.1000	-0.5564	0.4008	0.1200	0.6223	0.0800	-0.5275	0.4008	0.2300	0.6816	0.0000	-0.9684	-0.0470	0.0000	0.3441

Loss per epoch

training validation

Config

name: reference_bgl3_gcn_mr5_randomf1
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 6
  split_scheme: random
  split_fold: 1
  min_reads: 5
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