bgl3 · Graph CNN

reference_bgl3_gcn_mr5plm_contiguousf1_20261104-222540

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	1.2421	0.6123	0.6411	0.3280	151.0000	-0.0523	174.0000	0.0000	-0.1492	-0.0178	0.6324	0.2400	-0.2200	0.3347	0.0400	0.7040	0.2500	-0.2604	0.7630	0.1300	0.7322	0.1471	-0.2084	0.3302	0.0294	0.6849
validation	1.6812	0.3693	0.3131	0.1180	26.0000	-0.1446	664.0000	0.0000	-2.8339	-0.3377	-0.5222	0.0200	-2.1277	0.1847	0.0200	-0.2064	0.0100	-1.7886	0.1847	0.0600	-0.0211	0.0000	-2.7229	-0.3377	0.0000	-0.4744
test	1.4095	0.6295	0.6677	0.0177	40.0000	0.0648	254.0000	0.1000	-0.6600	0.0513	0.3440	0.1200	-0.4324	0.1987	0.1600	0.5913	0.1000	-0.5048	0.4008	0.2600	0.5872	0.0000	-0.7391	-0.1176	0.0000	0.2928

Loss per epoch

training validation

Config

name: reference_bgl3_gcn_mr5plm_contiguousf1
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  graph_threshold: 6
  split_scheme: contiguous
  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