Source code for grammar-based text-to-SQL parser using one variant of Transformer decoder called ASTormer. Implementations for prevalent cross-domain multi-table benchmarks Spider, SParC, CoSQL, DuSQL and Chase are all included.
Hello~ rhythm~ I want to ask why word level schema linking is used instead of using an subword level schema linking process.
I found a curious problem while reading the code. If PLM is used for tokenizing, would schema linking at the subword level be better than at the word level? Have you ever tried some tests about it? Thank you very much for your reply.
Because I found in LGESQL that there is indeed aggregation and pooling function at the subword token-level. So I am curious why RAT-SQL only needs word level. In addition, your lgesql is written very well. It is a rare and foundational article for the text-to-SQL task. Thank you very much for your contribution to the community.
Hello @rhythmcao, Do you have a previously trained model on the spider dataset, so that I can directly use it for inference. I'm only interested in using different metrics to check how different models perform. So it will be of real help if u can add the pretrained model directly useful for inference. Thanks in advance
I feel confused when reading the paper and code. I found that the encoder used in the paper is different from the one used in the code. Which one should I follow? Thank you.