ube4b · Graph CNN

reference_ube4b_gcn_modulof3_20261104-161644

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	3.9276	0.2734	0.1905	0.0064	7842.0000	0.3511	212.0000	0.6000	0.4066	5.9018	0.1904	0.5600	-0.0313	5.9018	0.0200	0.1646	0.6500	0.5103	5.9018	0.0200	0.2519	0.7291	1.2250	9.1537	0.1153	0.4084
validation	3.5112	-0.0141	-0.0471	-0.0144	318.0000	0.0365	381.0000	0.1000	-2.3402	0.2640	0.0000	-0.1300	0.0600	-2.4925	1.2902	0.0000	0.0000	-0.2142	0.1000	-2.3940	1.3191	0.0300	0.0000	-0.2241	0.1111	-2.4742	1.2902	0.0000	0.0000	-0.1719
test	3.0267	0.0830	0.0718	0.0006	418.0000	-0.0164	1235.0000	0.2000	-1.7733	0.2502	0.0000	-0.1012	0.2800	-1.3676	1.6642	0.0200	0.0000	-0.0189	0.2600	-1.3079	2.9172	0.0600	0.0000	-0.0028	0.3333	-1.1413	1.6642	0.0000	0.0000	0.0364

Loss per epoch

training validation

Config

name: reference_ube4b_gcn_modulof3
task: dms
data:
  dataset: ube4b
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 7
  split_scheme: modulo
  split_fold: 3
model:
  name: gcn
  filters:
  - 128
  - 128
  - 128
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7