donglixp / coarse2fine

Hi Li,
I understand the use of cond_span_l and con_span_r but doesn't the loss get calculated for these variables during training? Then why do we need cond_span_l_loss and cond_span_r_loss fields?

Also, is the variable len_tgt_list (declared in wikisql/table/IO.py line 181) useful? I see it declared but I cannot see it used anywhere.
Thank you,
Shruti

Hi
Could anyone tell me the PyTorch version of this project? I've got the error when running the code.

@KeSyren Please find the PyTorch version in https://github.com/donglixp/coarse2fine/blob/master/README.md .

@Sandy26

cond_span_l_loss and cond_span_r_loss are gold values that are used to compute loss during training.

coarse2fine/wikisql/table/Trainer.py

Line 112 in 8d135d0

    
           gold = {'agg': batch.agg, 'sel': batch.sel, 'lay': batch.lay, 'cond_col': batch.cond_col_loss,

@Sandy26

len_tgt_list is defined to get statistics about examples. It's not used during training and inference.

what is the use of "len_tgt_list" ,cond_span_l_loss,cond_span_r_loss?