bgl3 · Sequence CNN

reference_bgl3_cnn_plm_contiguousf3_20261105-040453

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	1.8333	0.3244	0.2953	0.0703	4815.0000	-0.0010	3124.0000	0.6000	0.5327	2.0174	0.2825	0.6000	0.4377	2.3136	0.0000	0.2896	0.6100	0.4752	2.3136	0.0100	0.3196	0.5942	0.4574	2.3136	0.0290	0.3224
validation	1.6526	0.2648	0.2028	0.0695	1997.0000	-0.1372	181.0000	0.5000	-0.0466	1.0480	0.0480	0.5400	-0.0222	2.0841	0.0000	0.0647	0.4300	-0.0836	2.0841	0.0200	0.0436	0.5319	-0.0583	2.0841	0.0000	0.0504
test	1.8274	0.3048	0.1881	0.0565	1899.0000	-0.0766	3049.0000	0.5000	-0.0009	1.5945	0.0887	0.4800	-0.2111	2.5002	0.0200	0.0260	0.5200	-0.0612	2.5002	0.0300	0.0858	0.4694	-0.2457	2.5002	0.0204	0.0134

Loss per epoch

training validation

Config

name: reference_bgl3_cnn_plm_contiguousf3
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: contiguous
  split_fold: 3
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