bgl3 · Graph CNN

reference_bgl3_gcn_randomf2_20261104-155213

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	precision_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	1.1934	0.6248	0.5858	0.3840	6114.0000	0.2089	3362.0000	1.0000	1.5085	3.1160	0.1000	0.5275	0.9800	1.3753	3.1160	0.0400	0.5578	0.9500	1.2332	3.1160	0.0400	0.5445	0.9313	1.2173	3.1160	0.0500	0.5629
validation	1.3808	0.5434	0.5037	0.2952	2110.0000	0.1129	476.0000	0.6000	0.5156	1.9459	0.0000	0.0000	0.2714	0.7200	0.7549	2.2054	0.0200	0.0000	0.3941	0.7800	0.8567	2.6204	0.0500	0.0000	0.4563	0.7358	0.7818	2.2054	0.0189	0.0000	0.4061
test	1.4435	0.5390	0.4935	0.2894	2044.0000	0.1278	2455.0000	0.8000	0.8601	1.6946	0.0000	0.0000	0.3906	0.7600	0.7449	1.9459	0.0000	0.0000	0.3973	0.7600	0.8115	2.5002	0.0400	0.0000	0.4498	0.7736	0.7703	1.9459	0.0000	0.0000	0.4087

Loss per epoch

training validation

Config

name: reference_bgl3_gcn_randomf2
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 6
  split_scheme: random
  split_fold: 2
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