gb1 · Graph CNN

reference_gb1_gcn_modulof0_20261104-144208

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	0.8733	0.9060	0.9122	0.8153	31059.0000	0.2518	5522.0000	1.0000	1.4761	1.9863	0.8415	0.9800	1.2456	1.9863	0.0200	0.8361	0.9900	1.2621	1.9863	0.0500	0.8583	0.9178	0.7606	2.0942	0.1936	0.8240
validation	5.7532	0.3599	0.3724	0.0079	4408.0000	0.2449	1062.0000	0.6000	-1.5254	1.1934	0.0000	0.2154	0.5600	-1.6837	1.2145	0.0400	0.0000	0.1842	0.6000	-1.5540	1.2145	0.0400	0.0000	0.2248	0.5628	-1.8024	1.2145	0.0603	0.0000	0.1600
test	4.9010	0.6842	0.6581	0.1219	2689.0000	0.1048	6802.0000	0.0000	-1.7981	-0.3621	0.0000	0.2419	0.1400	-0.9299	0.5783	0.0000	0.0000	0.4821	0.1700	-0.7162	0.5783	0.0000	0.0000	0.5516	0.1990	-1.0343	0.9804	0.0149	0.0000	0.4781

Loss per epoch

training validation

Config

name: reference_gb1_gcn_modulof0
task: dms
data:
  dataset: gb1
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 7
  split_scheme: modulo
  split_fold: 0
model:
  name: gcn
  filters:
  - 128
  - 128
  - 128
  - 128
  - 128
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 32
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7