How to import the models and predict on a fresh dataset? #7
Answered
by
YingtongDou
thepraveen19
asked this question in
Q&A
Discussed in https://github.com/safe-graph/GNN-FakeNews/discussions/5Originally posted by thepraveen19 December 24, 2021 Please help! |
Answered by
YingtongDou
Dec 29, 2021
Replies: 1 comment
|
Hi, to use the GNN models in this repo to classify a tweet, you need to build a tree-structured tweet propagation graph based on the retweets of the given tweets, which you can refer to the paper for more instructions. Once build the graph, you can use the pretrained model on the UPFD dataset to classify your tweets but I strongly suggest curating your own training data since there might be domain shifts for different news sets. |
0 replies
Answer selected by
YingtongDou
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi, to use the GNN models in this repo to classify a tweet, you need to build a tree-structured tweet propagation graph based on the retweets of the given tweets, which you can refer to the paper for more instructions.
Once build the graph, you can use the pretrained model on the UPFD dataset to classify your tweets but I strongly suggest curating your own training data since there might be domain shifts for different news sets.