bgl3 · Graph CNN

reference_bgl3_gcn_plm_contiguousf2_20261104-184418

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	1.6694	0.3386	0.2842	0.1057	4995.0000	-0.0669	2660.0000	0.6000	0.5672	2.0174	0.2724	0.5800	0.3901	2.2054	0.0000	0.2539	0.5200	0.1340	2.3136	0.0200	0.1779	0.5221	0.1529	2.3136	0.0147	0.1894
validation	1.7730	0.3018	0.1897	0.0846	1899.0000	-0.0812	3020.0000	0.6000	0.0811	1.5945	0.1145	0.5800	-0.0759	2.5002	0.0200	0.0749	0.4700	-0.3227	2.5002	0.0300	-0.0154	0.5918	-0.0379	2.5002	0.0204	0.0885
test	2.0775	0.2065	0.1133	0.0269	1465.0000	-0.0388	1597.0000	0.2000	-0.3476	0.3847	0.0452	0.4200	-0.3763	2.1466	0.0200	0.0413	0.3600	-0.4847	2.3136	0.0300	0.0037	0.4651	-0.1961	2.1466	0.0000	0.1027

Loss per epoch

training validation

Config

name: reference_bgl3_gcn_plm_contiguousf2
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  graph_threshold: 6
  split_scheme: contiguous
  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