gmm

Mixture Model Networks (GMM or MoNet)

• Paper link: https://arxiv.org/abs/1611.08402

Dataset Statics

Dataset	# Nodes	# Edges	# Classes
Cora	2,708	10,556	7
Pubmed	19,717	88,651	3

Refer to Planetoid.

Results

# available dataset: "cora", "pubmed"
TL_BACKEND="paddle" python gmm_trainer.py --dataset cora --n_epoch 100 --hidden_dim 32 --num_layers 2 --lr 0.01 --l2_coef 0.01 --drop_rate 0.8
TL_BACKEND="paddle" python gmm_trainer.py --dataset pubmed --n_epoch 200 --hidden_dim 32 --num_layers 2 --lr 0.01 --l2_coef 0.005 --drop_rate 0.9
TL_BACKEND="tensorflow" python gmm_trainer.py --dataset cora --n_epoch 200 --hidden_dim 32 --num_layers 2 --lr 0.01 --l2_coef 0.01 --drop_rate 0.8
TL_BACKEND="tensorflow" python gmm_trainer.py --dataset pubmed --n_epoch 200 --hidden_dim 32 --num_layers 2 --lr 0.01 --l2_coef 0.005 --drop_rate 0.9 
TL_BACKEND="torch" python gmm_trainer.py --dataset cora --n_epoch 200 --hidden_dim 32 --num_layers 2 --lr 0.007 --l2_coef 0.01 --drop_rate 0.8
TL_BACKEND="torch" python gmm_trainer.py --dataset pubmed --n_epoch 200 --hidden_dim 32 --num_layers 2 --lr 0.01 --l2_coef 0.01 --drop_rate 0.9
TL_BACKEND="mindspore" python gmm_trainer.py --dataset cora --n_epoch 100 --hidden_dim 32 --num_layers 2 --lr 0.01 --l2_coef 0.01 --drop_rate 0.8
TL_BACKEND="mindspore" python gmm_trainer.py --dataset pubmed --n_epoch 100 --hidden_dim 16 --num_layers 2 --lr 0.006 --l2_coef 0.005 --drop_rate 0.6

Dataset	Paper	Our(pd)	Our(tf)	Our(th)	Our(ms)
cora	81.69±0.48%	80.90±0.73%	81.68±0.65%	81.66±0.45%	80.06±1.19%
pubmed	78.81±0.44%	78.36±0.68%	79.08±0.53%	78.84±0.19%	78.08±0.39%