config ../config/ogbn-products/SDMP/ogbn-products_SDMP_base.yml
data ../dataset
result result/tmp
gnn result/cora/SAGE
device cuda:4
{'name': 'ogbn-products', 'epoch': 10, 'batch_size': 128, 'eval_step': 20, 'eval_batch_size': 128, 'feature_normalize': 'no', 'inductive_train': True, 'inductive_train_ratio': 0.01, 'theta_n_nonzero': 40, 'theta_cand_mode': 'mixed', 'theta_cand_k2': 3, 'theta_cand_fanout': [10, 10, 10], 'theta_cand_k1': 1, 'theta_cand_add_self': True, 'h_init_theta_mode': 'sparse', 'h_init_theta_k': 3, 'h_init_theta_k_fanout': [10, 10, 10], 'h_init_epoch': 1, 'h_hidden': [], 'h_loop_cnt': 1, 'h_lr': 0.001, 'h_l2': 1e-05, 'h_dropout': 0.0, 'target_h_model': 'SAGE'}
This will download 1.38GB. Will you proceed? (y/N)
Downloading https://snap.stanford.edu/ogb/data/nodeproppred/products.zip
config ../config/ogbn-products/SDMP/ogbn-products_SDMP_base.yml
data ../dataset
result result/tmp
gnn result/cora/SAGE
device cuda:4
{'name': 'ogbn-products', 'epoch': 10, 'batch_size': 128, 'eval_step': 20, 'eval_batch_size': 128, 'feature_normalize': 'no', 'inductive_train': True, 'inductive_train_ratio': 0.01, 'theta_n_nonzero': 40, 'theta_cand_mode': 'mixed', 'theta_cand_k2': 3, 'theta_cand_fanout': [10, 10, 10], 'theta_cand_k1': 1, 'theta_cand_add_self': True, 'h_init_theta_mode': 'sparse', 'h_init_theta_k': 3, 'h_init_theta_k_fanout': [10, 10, 10], 'h_init_epoch': 1, 'h_hidden': [], 'h_loop_cnt': 1, 'h_lr': 0.001, 'h_l2': 1e-05, 'h_dropout': 0.0, 'target_h_model': 'SAGE'}
This will download 1.38GB. Will you proceed? (y/N)
config ../config/ogbn-products/SDMP/ogbn-products_SDMP_base.yml
data ../dataset
result result/tmp
gnn result/cora/SAGE
device cuda:4
{'name': 'ogbn-products', 'epoch': 10, 'batch_size': 128, 'eval_step': 20, 'eval_batch_size': 128, 'feature_normalize': 'no', 'inductive_train': True, 'inductive_train_ratio': 0.01, 'theta_n_nonzero': 40, 'theta_cand_mode': 'mixed', 'theta_cand_k2': 3, 'theta_cand_fanout': [10, 10, 10], 'theta_cand_k1': 1, 'theta_cand_add_self': True, 'h_init_theta_mode': 'sparse', 'h_init_theta_k': 3, 'h_init_theta_k_fanout': [10, 10, 10], 'h_init_epoch': 1, 'h_hidden': [], 'h_loop_cnt': 1, 'h_lr': 0.001, 'h_l2': 1e-05, 'h_dropout': 0.0, 'target_h_model': 'SAGE'}
This will download 1.38GB. Will you proceed? (y/N)
config ../config/ogbn-products/SDMP/ogbn-products_SDMP_base.yml
data ../dataset
result result/tmp
gnn result/cora/SAGE
device cuda:4
{'name': 'ogbn-products', 'epoch': 10, 'batch_size': 128, 'eval_step': 20, 'eval_batch_size': 128, 'feature_normalize': 'no', 'inductive_train': True, 'inductive_train_ratio': 0.01, 'theta_n_nonzero': 40, 'theta_cand_mode': 'mixed', 'theta_cand_k2': 3, 'theta_cand_fanout': [10, 10, 10], 'theta_cand_k1': 1, 'theta_cand_add_self': True, 'h_init_theta_mode': 'sparse', 'h_init_theta_k': 3, 'h_init_theta_k_fanout': [10, 10, 10], 'h_init_epoch': 1, 'h_hidden': [], 'h_loop_cnt': 1, 'h_lr': 0.001, 'h_l2': 1e-05, 'h_dropout': 0.0, 'target_h_model': 'SAGE'}
This will download 1.38GB. Will you proceed? (y/N)
Downloading https://snap.stanford.edu/ogb/data/nodeproppred/products.zip
Extracting dataset/products.zip
Loading necessary files...
This might take a while.
Processing graphs...
Converting graphs into DGL objects...
Saving...
Refined train size: 196615, val size: 39323, test size: 2213091.
Loaded theta from ../dataset/ogbn-products/SDMPPre/SAGE/theta_mixed_k1_1_k2_3_10_10_10_withself.
Loaded h_init_theta from ../dataset/ogbn-products/SDMPPre/SAGE/h_init_theta_sparse_3_10_10_10.
Loaded X from ../dataset/ogbn-products/SDMPPre/SAGE/features_normed_no.
Loaded target from ../dataset/ogbn-products/SDMPPre/SAGE/target.
Process SAGE finished in 34.5 s.
Initializing the model...
