TimDettmers / ConvE

Convolutional 2D Knowledge Graph Embeddings resources
MIT License
675 stars 163 forks source link

Can I use more network layers? #79

Open quqxui opened 3 years ago

quqxui commented 3 years ago

Hi, thanks for your job and code, I have a question about that the number of network layer. I noticed you only using one conv2d and one FC, as everyone knows, deep model can get better performance. I wonder why you don't use more convolution and FC layers? Doesn't the deep model help performance?

Hope your reply, thanks again.