TaoranJ / DMA-Nets

The official Pytorch implementation of paper "Dynamic Multi-Context Attention Networks for Citation Forecasting of Scientific Publications"
MIT License
2 stars 0 forks source link

A question about data #4

Open FairyLee-web opened 11 months ago

FairyLee-web commented 11 months ago

Hello, I am very interested in your research. I noticed that there is data for inventors and assignees in the dataset, but it doesn't seem to be reflected in the code. Is it correct that the DMA model you proposed only uses the data from the patents themselves?

TaoranJ commented 8 months ago

Hi, thanks for asking. DMA-Nets only consider one time series so in both training and evaluation, assignee and inventor time series were dropped.