gb1 · Sequence CNN

reference_gb1_cnn_contiguousf0_20261104-144946

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	0.2640	0.9803	0.9605	0.9549	32275.0000	0.9140	1166.0000	1.0000	1.6207	1.9362	0.8918	1.0000	1.7061	2.0942	0.1000	0.9454	1.0000	1.6846	2.1952	0.2600	0.9577	1.0000	1.4132	2.4539	0.7848	0.9898
validation	2.1670	-0.0508	-0.0428	-0.0738	5403.0000	0.1082	3818.0000	0.8000	-0.5978	0.5124	0.0000	0.1760	0.5400	-0.7288	0.8474	0.0000	0.0000	0.1303	0.4800	-0.8358	0.8474	0.0000	0.0000	0.0846	0.4577	-0.7980	1.0909	0.0100	0.0000	0.1076
test	3.8889	-0.1434	-0.2084	-0.2669	1866.0000	-0.1368	17937.0000	0.0000	-3.4207	-1.7131	0.0000	-0.4632	0.0000	-2.6908	-1.4760	0.0000	0.0000	-0.2429	0.0000	-2.6514	-1.1274	0.0000	0.0000	-0.2355	0.0000	-2.6549	-0.8634	0.0000	0.0000	-0.2440

Loss per epoch

training validation

Config

name: reference_gb1_cnn_contiguousf0
task: dms
data:
  dataset: gb1
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: contiguous
  split_fold: 0
model:
  name: cnn
  conv:
  - - 128
    - 17
  - - 128
    - 17
  - - 128
    - 17
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 32
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7