pab1 · Graph CNN

reference_pab1_gcn_contiguousf2_20261104-164436

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	precision_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	1.0989	0.8489	0.8610	0.6993	1747.0000	0.0913	6089.0000	0.5000	0.1769	0.8662	0.2000	0.7224	0.5200	0.0442	0.8662	0.0400	0.7413	0.4900	-0.0371	0.8662	0.0400	0.7431	0.4833	-0.0279	0.8662	0.0944	0.7749
validation	2.6470	-0.0849	-0.0959	-0.0431	383.0000	0.0075	1519.0000	0.0000	-2.5576	-1.2185	0.0000	-0.6818	0.0200	-3.2865	0.1586	0.0000	-1.2023	0.0200	-3.0186	0.5248	0.0100	-1.1354	0.0000	-3.2712	-0.1367	0.0000	-1.1441
test	6.0737	-0.3842	-0.3693	-0.4300	475.0000	0.1335	611.0000	0.0000	-5.1925	-4.3322	0.0000	-1.0001	0.0000	-4.9741	-0.8578	0.0000	-0.9947	0.0100	-4.8232	0.5486	0.0100	-0.9728	0.0000	-5.0049	-0.8578	0.0000	-1.0041

Loss per epoch

training validation

Config

name: reference_pab1_gcn_contiguousf2
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 7
  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