search

dongyuanjushi / LightLM

22 stars 2 forks source link

Preprocessing ML-1m dataset for Graph Collaborative Indexing (CoUI) #2

Open woriazzc opened 8 months ago

woriazzc commented 8 months ago

Hi! It's a great work!

I'm trying to run Graph Collaborative Indexing (CoUI) on ML-1m dataset. I first run GID_generation.py. But it needs the "BPRMF_user_emb_64.npy". Should I write extra codes to generate it? Or should I run some other files first?

dongyuanjushi commented 8 months ago

Thank you for your attention to our work, to perform graph collaborative indexing in this version of our code, you need to get the embedding vector from the BPRMF algorithm, which is based on https://github.com/THUwangcy/ReChorus.git. In the BPRMF, you can setup embedding dim as 64 and save the embedding vectors. And we are optimizing the code to perform the graph collaborative indexing, which will be updated soon.