bgl3 · Graph CNN

reference_bgl3_gcn_contiguousf0_20261104-144925

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	1.9631	0.4586	0.4283	0.0129	6904.0000	0.0659	13298.0000	0.7000	0.8391	2.0841	0.3361	0.6200	0.3960	2.3637	0.0000	0.2240	0.6200	0.3670	3.1160	0.0300	0.2267	0.6023	0.3588	3.1160	0.0409	0.2345
validation	1.6590	-0.0989	-0.0859	-0.0015	1203.0000	-0.0130	1157.0000	0.1000	-1.9981	1.3581	0.0000	-0.6091	0.1200	-1.6765	1.3581	0.0000	0.0000	-0.5380	0.2100	-1.2564	1.9459	0.0100	0.0000	-0.3986	0.0882	-1.7144	1.3581	0.0000	0.0000	-0.5363
test	2.1030	-0.0146	-0.0080	-0.0494	1218.0000	-0.0023	876.0000	0.0000	-2.0230	-0.2513	0.0000	-0.3835	0.1000	-1.6081	1.2150	0.0000	0.0000	-0.3124	0.1700	-1.2787	1.3581	0.0000	0.0000	-0.2174	0.0930	-1.6942	1.2150	0.0000	0.0000	-0.3361

Loss per epoch

training validation

Config

name: reference_bgl3_gcn_contiguousf0
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 6
  split_scheme: contiguous
  split_fold: 0
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