bgl3 · Graph CNN

reference_bgl3_gcn_mr5plm_modulof1_20261104-222504

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	0.6085	0.8034	0.8103	0.6171	136.0000	0.0424	585.0000	0.3000	-0.1259	0.5578	0.1000	0.6299	0.3200	-0.1204	0.5578	0.1600	0.7296	0.2600	-0.1360	0.5578	0.1800	0.7758	0.3333	-0.1314	0.5578	0.1111	0.6792
validation	0.8967	0.4882	0.5719	0.2368	23.0000	0.0539	9.0000	0.3000	-0.6396	0.3940	0.1000	0.3396	0.1800	-0.3274	0.3940	0.2600	0.6713	0.1400	-0.4584	0.3940	0.3300	0.6024	0.1250	-0.8521	0.0184	0.0000	0.1807
test	0.7583	0.5989	0.6016	0.3281	47.0000	0.0071	6.0000	0.1000	-0.3664	1.6908	0.1000	0.4040	0.1000	-0.4659	1.6908	0.1000	0.4490	0.1500	-0.3968	1.6908	0.2500	0.5653	0.1429	-0.2823	1.6908	0.1429	0.4271

Loss per epoch

training validation

Config

name: reference_bgl3_gcn_mr5plm_modulof1
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  graph_threshold: 6
  split_scheme: modulo
  split_fold: 1
  min_reads: 5
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