bgl3 · Graph CNN

reference_bgl3_gcn_mr5_modulof1_20261104-221918

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	0.7338	0.7767	0.7741	0.5384	136.0000	0.0094	1278.0000	0.2000	-0.0789	0.3347	0.6540	0.2400	-0.2177	0.5578	0.1000	0.6720	0.2400	-0.2263	0.5578	0.1800	0.7182	0.2593	-0.1858	0.3347	0.0741	0.6490
validation	1.1385	0.1961	0.1282	0.0309	23.0000	0.0726	17.0000	0.0000	-1.5083	-0.0887	-0.2940	0.0400	-1.3013	0.3940	0.0400	-0.1693	0.0400	-1.0928	0.3940	0.1300	0.0115	0.0000	-1.7269	-0.0887	0.0000	-0.4439
test	1.0732	0.3083	0.2429	0.0491	47.0000	-0.0870	21.0000	0.1000	-1.2430	0.2216	-0.1744	0.1400	-1.1957	1.6908	0.1400	-0.1899	0.1000	-1.0839	1.6908	0.1700	-0.1020	0.1429	-0.9003	0.2216	0.0000	0.0481

Loss per epoch

training validation

Config

name: reference_bgl3_gcn_mr5_modulof1
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 6
  split_scheme: modulo
  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