avgfp · Fully connected

reference_avgfp_fc_modulof0_20261104-144009

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	0.2830	0.8706	0.7748	0.6820	1210.0000	0.1879	142.0000	0.6000	3.7168	3.8486	0.6744	0.4200	3.6785	4.0088	0.0200	0.6915	0.4100	3.6814	4.0860	0.0800	0.7260	0.4255	3.6840	4.0860	0.0851	0.7259
validation	0.5409	-0.2972	-0.2420	-0.0706	216.0000	0.0162	1049.0000	0.0000	2.0482	3.4446	0.0000	-1.8966	0.1600	2.5327	3.7978	0.0000	0.0000	-1.4275	0.1600	2.7321	3.8543	0.0800	0.0000	-1.1557	0.0000	2.0080	3.4446	0.0000	0.0000	-1.9812
test	0.6012	-0.2828	-0.2537	-0.0580	146.0000	0.1990	61.0000	0.0000	2.0129	3.7002	0.0000	-2.0386	0.0600	2.6067	3.8057	0.0200	0.0000	-1.3141	0.1400	2.9001	4.0192	0.0900	0.0000	-0.8309	0.0000	1.9374	3.4444	0.0000	0.0000	-2.0939

Loss per epoch

training validation

Config

name: reference_avgfp_fc_modulof0
task: dms
data:
  dataset: avgfp
  encoding: one_hot,aa_index
  split_dir: null
  split_scheme: modulo
  split_fold: 0
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