avgfp · Graph CNN

reference_avgfp_gcn_contiguousf1_20261104-153940

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	0.5330	0.5928	0.4983	0.2206	1898.0000	0.0513	3937.0000	0.1000	3.6387	3.8727	0.5036	0.3200	3.6472	3.8984	0.0000	0.5553	0.3300	3.6105	3.9031	0.0000	0.5360	0.3091	3.5942	3.9031	0.0091	0.5186
validation	0.4864	0.4216	0.2577	0.0254	142.0000	-0.0667	110.0000	0.1000	2.8696	3.7277	0.0000	-0.7618	0.0800	3.1351	3.7366	0.0000	0.0000	-0.3698	0.1000	3.3161	3.8359	0.0300	0.0000	-0.0193	0.0909	2.9051	3.7277	0.0000	0.0000	-0.7088
test	1.1824	0.3748	0.3078	-0.1761	60.0000	-0.1589	15.0000	0.0000	2.6985	3.6630	0.0000	-0.1287	0.2000	3.1295	3.9923	0.1800	0.0000	0.3087	0.1400	3.1064	3.9923	0.2100	0.0000	0.2977	0.0000	2.9146	3.3346	0.0000	0.0000	0.0725

Loss per epoch

training validation

Config

name: reference_avgfp_gcn_contiguousf1
task: dms
data:
  dataset: avgfp
  encoding: one_hot,aa_index
  split_dir: null
  graph_threshold: 7
  split_scheme: contiguous
  split_fold: 1
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