bgl3 · Graph CNN

reference_bgl3_gcn_mr5plm_contiguousf0_20261104-222340

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	1.1138	0.7168	0.7486	0.4341	156.0000	0.0457	36.0000	0.4000	-0.0917	0.3940	0.6602	0.4200	-0.0278	1.6908	0.2000	0.8045	0.2900	-0.1145	1.6908	0.2100	0.8140	0.3429	-0.1077	0.3940	0.1714	0.7397
validation	0.8526	0.6407	0.6719	0.4058	40.0000	0.0963	229.0000	0.1000	-0.3061	0.1090	0.5511	0.1600	-0.3919	0.4008	0.1800	0.6206	0.1100	-0.3952	0.4008	0.2800	0.6738	0.1111	-0.3155	0.1090	0.0000	0.5365
test	1.5778	0.5493	0.5665	0.0708	21.0000	0.0448	164.0000	0.0000	-1.5460	-0.3598	0.1312	0.0400	-1.0586	0.2745	0.1200	0.4099	0.0600	-0.9402	0.4991	0.1900	0.5052	0.0000	-1.5145	-0.3598	0.0000	0.1479

Loss per epoch

training validation

Config

name: reference_bgl3_gcn_mr5plm_contiguousf0
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  graph_threshold: 6
  split_scheme: contiguous
  split_fold: 0
  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