pab1 · Graph CNN

reference_pab1_gcn_contiguousf4_20261104-170958

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	1.8021	0.7626	0.7715	0.5663	1652.0000	0.0080	865.0000	0.2000	-0.1966	0.3298	0.6469	0.2400	-0.1393	0.6979	0.0600	0.7165	0.2100	-0.1865	0.6979	0.0400	0.7250	0.2556	-0.1739	0.7097	0.0556	0.7521
validation	4.0141	0.1074	0.0889	-0.0829	222.0000	0.0745	2519.0000	0.0000	-2.6682	-0.2024	-0.0202	0.0200	-2.7779	0.0176	0.0000	-0.0585	0.0200	-2.4802	0.0644	0.0000	0.0440	0.0204	-2.8141	0.0176	0.0000	-0.0707
test	2.9417	-0.2539	-0.2610	-0.5770	931.0000	0.0848	3559.0000	0.0000	-2.8397	-0.5933	-0.9055	0.0400	-2.5256	0.2711	0.0000	-0.8667	0.0600	-2.3248	0.3585	0.0000	-0.8125	0.0444	-2.6576	0.2711	0.0000	-0.9317

Loss per epoch

training validation

Config

name: reference_pab1_gcn_contiguousf4
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 7
  split_scheme: contiguous
  split_fold: 4
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