pab1 · Sequence CNN

reference_pab1_cnn_modulof0_20261104-144559

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	2.5350	0.5910	0.5539	0.3153	1755.0000	0.0544	6645.0000	0.2000	-0.3305	0.3458	0.5209	0.1800	-0.4416	0.4410	0.0000	0.5254	0.1600	-0.4539	0.5930	0.0100	0.5442	0.1522	-0.5044	0.5930	0.0217	0.5343
validation	4.2187	0.2536	0.2790	-0.0569	132.0000	0.1728	458.0000	0.1000	-0.9919	0.2996	0.0000	0.4351	0.1800	-1.3000	0.4681	0.1200	0.0000	0.3692	0.1700	-1.2428	0.4893	0.1400	0.0000	0.4094	0.2353	-0.8594	0.4681	0.1176	0.0000	0.5025
test	3.2513	-0.0303	-0.0172	-0.0121	171.0000	0.0782	132.0000	0.0000	-4.2682	-0.4963	0.0000	-0.9739	0.0400	-3.1605	0.1694	0.0000	0.0000	-0.6202	0.0800	-2.5782	0.3875	0.0700	0.0000	-0.3701	0.0000	-3.7141	-0.1128	0.0000	0.0000	-0.7964

Loss per epoch

training validation

Config

name: reference_pab1_cnn_modulof0
task: dms
data:
  dataset: pab1
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: modulo
  split_fold: 0
model:
  name: cnn
  conv:
  - - 128
    - 17
  - - 128
    - 17
  - - 128
    - 17
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 128
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7