search

chennnM / GCNII

PyTorch implementation of "Simple and Deep Graph Convolutional Networks"
325 stars 72 forks source link

Questions about 100 runs #5

Closed tpy1990wy closed 4 years ago

tpy1990wy commented 4 years ago

Great work！We are reproducing the statistical results in Table 2 and want to know how to design the 100 runs mentioned in the article for GCN and GCNII. Use random seeds and fixed hyperparameters in Table 6? Or some other method to select 100 runs?

chennnM commented 4 years ago

Hi, I used the hyperparameters in Table 6, commented out the code for setting seed, and repeated the network initialization and training 100 times.

tpy1990wy commented 4 years ago

@chennnM Thank you for your reply.