avgfp · Fully connected

reference_avgfp_fc_contiguousf3_20261104-161729

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	0.3482	0.8318	0.7894	0.6643	805.0000	0.1875	56.0000	0.2000	3.6612	3.9226	0.6919	0.2800	3.6732	4.0860	0.0600	0.7511	0.4100	3.6959	4.1231	0.1100	0.8001	0.3810	3.6919	4.1231	0.0714	0.7888
validation	0.4256	-0.1188	-0.0539	-0.0391	156.0000	0.0526	68.0000	0.1000	2.5486	3.7268	0.0000	-1.7389	0.1000	2.9462	3.8060	0.0200	0.0000	-1.0522	0.1500	3.1586	3.9813	0.1000	0.0000	-0.6062	0.1250	2.2821	3.7268	0.0000	0.0000	-2.2505
test	0.2586	-0.1281	-0.1644	-0.0922	366.0000	0.0361	802.0000	0.1000	3.2240	3.7919	0.0000	-0.5739	0.2400	3.2808	3.8790	0.0200	0.0000	-0.6087	0.1900	3.3536	4.0306	0.0500	0.0000	-0.4848	0.0769	3.2885	3.7919	0.0000	0.0000	-0.4682

Loss per epoch

training validation

Config

name: reference_avgfp_fc_contiguousf3
task: dms
data:
  dataset: avgfp
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: contiguous
  split_fold: 3
model:
  name: fcn
  hidden:
  - 100
  - 100
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 32
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7