pab1 · Graph CNN

reference_pab1_gcn_plm_modulof0_20261104-175202

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	0.8521	0.8779	0.8365	0.7698	1755.0000	0.0452	1660.0000	0.1000	-0.2296	0.0979	0.5573	0.3800	-0.1169	0.8601	0.0600	0.6532	0.3600	-0.1493	0.8908	0.0600	0.6696	0.3406	-0.1520	0.8908	0.0725	0.6823
validation	1.8167	0.7387	0.7409	0.5449	132.0000	-0.2611	570.0000	0.1000	-0.2394	0.0189	0.6986	0.1600	-0.2490	0.3930	0.0200	0.7843	0.1500	-0.2588	0.3930	0.1000	0.8156	0.1176	-0.2567	0.0644	0.0000	0.7228
test	1.4712	0.7910	0.7819	0.5420	171.0000	0.0207	383.0000	0.0000	-0.2653	-0.0313	0.6187	0.2000	-0.2165	0.2375	0.0400	0.7424	0.2600	-0.2098	0.9033	0.1700	0.7865	0.1176	-0.2186	0.2232	0.0000	0.6737

Loss per epoch

training validation

Config

name: reference_pab1_gcn_plm_modulof0
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  graph_threshold: 7
  split_scheme: modulo
  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