bgl3 · Fully connected

reference_bgl3_fc_mr5plm_modulof2_20261104-222657

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	precision_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	1.1402	0.5761	0.6101	0.2846	143.0000	0.0380	328.0000	0.2000	-0.6983	0.2173	0.0000	0.3056	0.0800	-0.6076	0.2173	0.0200	0.4311	0.1200	-0.5865	0.7630	0.0700	0.4809	0.1111	-0.5379	0.2173	0.0000	0.4365
validation	0.7358	0.5172	0.4902	0.2663	44.0000	0.0166	332.0000	0.1000	-0.4694	0.3961	0.1000	0.3250	0.1400	-0.5596	0.3961	0.1400	0.3305	0.1400	-0.5511	0.3961	0.2900	0.3739	0.1429	-0.5382	0.3961	0.1429	0.2611
test	0.9530	0.4510	0.5149	0.1888	23.0000	-0.2342	136.0000	0.0000	-1.5192	-0.0455	0.0000	-0.3020	0.0600	-0.9466	0.1421	0.1200	0.1369	0.0700	-0.7050	0.1448	0.3000	0.3727	0.0000	-1.5283	-0.0455	0.0000	-0.3021

Loss per epoch

training validation

Config

name: reference_bgl3_fc_mr5plm_modulof2
task: dms
data:
  dataset: bgl3
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  split_scheme: modulo
  split_fold: 2
  min_reads: 5
model:
  name: fcn
  hidden:
  - 100
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 32
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7