Dataset format - Githubissues

jma127 / pyltr

Python learning to rank (LTR) toolkit

BSD 3-Clause "New" or "Revised" License

463 stars 107 forks source link

Open Jason-WT opened 5 years ago

Jason-WT commented 5 years ago

What format does the model need to train and test? (Only the format of LETOR can be used in the model??)

KasSanderink commented 4 years ago

Using the following seems to work for me:

TX: a pandas dataframe containing all features (except the target values and the group ids)
Ty: a pandas series containing the target values
Tqids: a pandas series (same length as Ty) cotaining the group id for each instance

Using numpy arrays instead of dataframes probably works as well