search

acbull / pyHGT

Code for "Heterogeneous Graph Transformer" (WWW'20), which is based on pytorch_geometric
MIT License
775 stars 162 forks source link

About Dataset #56

Open DuanhaoranCC opened 1 year ago

DuanhaoranCC commented 1 year ago

Hello! Thanks for your solid work, I have a couple of dataset questions.

  1. CS and Med are a subset of OAG, and graph_CS is a subset of CS. So what is the relationship between ML, NN and graphCS?
  2. graph_CS has the following relationship: ('paper', 'PV_Conference', 'avenue'), ('paper', 'PV_Repository', 'avenue'), ('paper', 'PV_Patent', 'avenue'), ('paper', 'PP_cite', 'paper'), (' author', 'AP_write_last', 'paper'). ('author', 'AP_write_other', 'paper'), ('author', 'AP_write_first', 'paper'), ('field', 'FF_in', 'field'), ('paper', 'PF_in_L0', 'field'), (' paper', 'PF_in_L3', 'field'), ('paper', 'PF_in_L1', 'field'), ('paper', 'PF_in_L2', 'field'), ('paper', 'PF_in_L5', 'field'), ('paper', 'PF_in_L4', 'field'), ('paper', 'PF_in_L5', 'field'), ('paper', 'PF_in L4', 'field'), ('author', 'in', 'affiliation')] What do they mean? ('PV_Conference', 'PV_Repository', 'PV_Patent', 'AP_write_last', 'AP_write_other',)
  3. What does the -1 in the label stand for?