avgfp · Graph CNN

reference_avgfp_gcn_plm_modulof4_20261105-042231

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	0.3629	0.7876	0.7136	0.5984	1414.0000	0.1562	1749.0000	0.4000	3.6802	3.9081	0.0000	0.6379	0.3200	3.6634	3.9081	0.0000	0.6749	0.3600	3.6587	3.9081	0.0300	0.7002	0.3578	3.6580	3.9081	0.0275	0.7035
validation	0.5008	0.3813	0.3661	0.1188	146.0000	-0.1048	468.0000	0.3000	3.4327	3.7468	0.0000	0.1929	0.2800	3.4531	3.9546	0.0800	0.2675	0.2500	3.5257	4.0087	0.1400	0.4375	0.1429	3.3159	3.7315	0.0000	0.0091
test	0.4840	0.2135	0.2799	0.0348	152.0000	-0.0278	346.0000	0.3000	3.1570	3.8918	0.1000	-0.2864	0.1400	3.3312	3.9912	0.0600	0.0019	0.2100	3.4101	3.9912	0.1300	0.1692	0.2500	3.0281	3.8918	0.0000	-0.4907

Loss per epoch

training validation

Config

name: reference_avgfp_gcn_plm_modulof4
task: dms
data:
  dataset: avgfp
  encoding: one_hot,aa_index,plm,plm_sum
  split_dir: null
  graph_threshold: 7
  split_scheme: modulo
  split_fold: 4
model:
  name: gcn
  filters:
  - 128
  - 128
  hidden:
  - 100
  dropout: 0.2
train:
  epochs: 300
  batch_size: 32
  learning_rate: 0.0001
  patience: 15
  min_delta: 1.0e-05
  seed: 7