Using GAT for regression

Diego999 / pyGAT

Pytorch implementation of the Graph Attention Network model by Veličković et. al (2017, https://arxiv.org/abs/1710.10903)

MIT License

2.84k stars 685 forks source link

Using GAT for regression #43

Open AdrianBZG opened 3 years ago

AdrianBZG commented 3 years ago

Hi,

I would like to try this model out in a regression scenario, what sould be changed in order to do so? Is using a different loss function (one for regression) enough?

Thanks.

tibe97 commented 3 years ago

I think you should also remove the softmax before returning the output in the forward method

SuperBruceJia commented 3 years ago

How about the labels w.r.t. the regression problem? @Diego999 @tibe97 @AdrianBZG @pbloem @sh0416