In this issue page, we will report the challenges through refactoring steps of SEERa and mostly for the graph embedding layer. Multiple steps can be defined for this task as listed below (the list is going to be updated regularly):
adding (followship only/ latent only/ followship and latent) options
extracting followship information out of the data
adding topics as node features (if the latent option is on)
creating edges as node connections (followship or latent based on the user choice)
predict node features to obtain user interest in the future
predict edges to obtain users' connections in the future
preparing the dataset to be fed to pyg temporal nn models
removing dynamicgem and using pyg temporal for graph embedding
check if sentence transformers can embed raw strings better than simple BOW https://www.sbert.net/
Which loss functions are suitable for which problems?
In this issue page, we will report the challenges through refactoring steps of SEERa and mostly for the graph embedding layer. Multiple steps can be defined for this task as listed below (the list is going to be updated regularly):