Dataset format
Jason-WT opened this issue · comments
What format does the model need to train and test? (Only the format of LETOR can be used in the model??)
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