Abhijeet Awasthi

> Okay, Do @awasthiabhijeet the spider dataset, train_spider.json, each question contains few keys, like query, question , sql, etc so what is that sql is exactly,what it contains, and is...

Hi @OhadRubin , I'm adding the following assertion below line [682](https://github.com/OhadRubin/SmBop/blob/e7a6fce7af5aa5545bd3cfca6c4c4dbef610cd6b/smbop/models/smbop.py#L682): ``` check_is_levelorder_list = is_levelorder_list.sum(-1) check_is_levelorder_list = (check_is_levelorder_list > 0).float() check_is_levelorder_list = check_is_levelorder_list.prod() assert check_is_levelorder_list > 0 ``` The assertion...

Thanks for the clarification, @OhadRubin Apart from the "French" -- "France" example, where else could teacher forcing fail? At each level, shouldn't the following line ensure that gold trees are...

Hi @OhadRubin , On the Spider train set, I observe that `final_beam_acc` with teacher forcing enabled, is more than 96% But, on the Spider val set, `final_beam_acc` with teacher forcing...

Sorry, there was a mistake in how I was applying teaching forcing on the val set. It seems to be working fine, with `final_beam_acc` close to 96% on the val...

Hi @tscholak, Thanks for open-sourcing this project. Starting from pre-trained checkpoints would be very helpful. Have there been any updates around the pre-trained checkpoints? Thanks,