model:
  d_model: 32
  d_state: 2
  d_conv: 4
  expand: 2
  n_layers: 1
  dropout: 0.5

data:
  batch_size: 32
  max_nodes: 10000
  test_split: 0.2
  
training:
  learning_rate: 0.01
  weight_decay: 0.0005
  epochs: 200
  patience: 20
  warmup_epochs: 10
  min_lr: 1e-6

ordering:
  strategy: "bfs"  # bfs, spectral, degree, community
  preserve_locality: true

evaluation:
  metrics: ["accuracy", "f1_macro", "f1_micro"]
  save_best: true