gb1 · Fully connected

reference_gb1_fc_modulof3_20261104-161045

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	0.3767	0.9693	0.9628	0.9302	29512.0000	0.7904	829.0000	1.0000	1.3063	1.4526	0.9144	1.0000	1.2266	1.4526	0.2200	0.9502	1.0000	1.1723	1.4526	0.3100	0.9557	0.9990	0.8843	1.8705	0.6109	0.9737
validation	4.4812	0.2038	0.2068	0.0302	3230.0000	0.0078	4073.0000	0.3000	-0.6454	0.2940	0.0000	0.4268	0.2600	-0.5088	0.5697	0.0000	0.0000	0.4804	0.2200	-0.7195	0.8659	0.0000	0.0000	0.4324	0.2488	-0.7388	1.1206	0.0050	0.0000	0.4387
test	4.6984	-0.1064	-0.1278	-0.0712	4705.0000	0.0643	11629.0000	0.0000	-1.6576	-0.4108	0.0000	0.0633	0.0200	-1.7528	0.0377	0.0000	0.0000	0.0425	0.0900	-1.9258	0.7183	0.0000	0.0000	-0.0039	0.1500	-1.7699	1.7376	0.0050	0.0000	0.0400

Loss per epoch

training validation

Config

name: reference_gb1_fc_modulof3
task: dms
data:
  dataset: gb1
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: modulo
  split_fold: 3
model:
  name: fcn
  hidden:
  - 1000
  dropout: 0.2
train:
  epochs: 300
  batch_size: 64
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7