expressgnn's Introduction

ExpressGNN

This is an implementation of the ExpressGNN proposed in the paper "Efficient Probabilistic Logic Reasoning with Graph Neural Networks".

Requirements

python 3.7
pytorch 1.1
scikit-learn
networkx
tqdm

Quick Start

The following command starts the inference on the Kinship-S1 dataset on GPU:

python -m main.train -data_root data/kinship/S1 -slice_dim 8 -batchsize 16 -use_gcn 1 -embedding_size 64 -gcn_free_size 32 -load_method 0 -exp_folder exp -exp_name kinship -device cuda

To run ExpressGNN on the FB15K-237 dataset on GPU, use the follwoing command line: Free Parameters + GNN, Free parameters dim = free_dim, GNN parameters dim = embedding_size - free_dim

python -m main.train -slice_dim 16 -batchsize 16 -num_hops 1 -patience 20 -lr_decay_patience 100 -entropy_temp 1 -load_method 1 -exp_folder exp -device cuda -rule_filename cleaned_rules_weight_larger_than_0.9.txt -data_root data/fb15k-237 -embedding_size 128 -gcn_free_size 127  -exp_name fb15k-237 -mylambda 1 -exp_mode 1

GNN parameters, Free parameters dim = 0, GNN parameters dim = embedding_size

python -m main.train -slice_dim 16 -batchsize 16 -num_hops 1 -patience 20 -lr_decay_patience 100 -entropy_temp 1 -load_method 1 -exp_folder exp -device cuda -rule_filename cleaned_rules_weight_larger_than_0.9.txt -data_root data/fb15k-237 -embedding_size 128 -exp_name fb15k-237 -mylambda 1 -exp_mode 2

GNN parameters, Free parameters dim = embedding_size, GNN parameters dim = 0

python -m main.train -slice_dim 16 -batchsize 16 -num_hops 1 -patience 20 -lr_decay_patience 100 -entropy_temp 1 -load_method 1 -exp_folder exp -device cuda -rule_filename cleaned_rules_weight_larger_than_0.9.txt -data_root data/fb15k-237 -embedding_size 128 -exp_name fb15k-237 -mylambda 1 -exp_mode 3

Recommend Projects