bgl3 · Sequence CNN

reference_bgl3_cnn_plm_contiguousf0_20261104-175419

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	1.6729	0.4498	0.3977	0.1588	6904.0000	0.0241	8043.0000	0.5000	0.3648	2.2054	0.1932	0.6200	0.5956	2.5419	0.0200	0.2905	0.6700	0.6541	2.5419	0.0300	0.3277	0.7193	0.7065	3.1160	0.0468	0.3628
validation	1.4873	0.3270	0.2769	0.1021	1203.0000	-0.0096	1913.0000	0.3000	-0.4045	1.5945	-0.0303	0.3800	-0.3222	1.5945	0.0000	-0.0004	0.3400	-0.2409	1.5945	0.0000	0.0342	0.3235	-0.4396	1.5945	0.0000	-0.0456
test	1.9971	0.3190	0.2849	0.0034	1218.0000	-0.0597	843.0000	0.5000	-0.3479	1.0109	0.0941	0.3400	-0.4508	1.4833	0.0000	0.0762	0.3500	-0.3267	1.7856	0.0100	0.1271	0.3256	-0.4506	1.4833	0.0000	0.0752

Loss per epoch

training validation

Config

name: reference_bgl3_cnn_plm_contiguousf0
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: contiguous
  split_fold: 0
model:
  name: cnn
  conv:
  - - 32
    - 17
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 64
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7