Open dnaihao opened 3 years ago
Yes, this is noted in Step 3 of the "SMCalFlow Experiments" section in the README.md.
We use --include_described_entities to add the entities (e.g., entity@123456) described in the generation outcome for the context turns. These entities mentioned in the context turns can appear in the "inlined" programs for the current turn, and thus, we include them in the source sequence so that the seq2seq model can produce such tokens via a copy mechanism.
Does this answer your question?
Thanks for the information! But what about the number after the symbol @? Is there a pattern for that?
The number after the symbol is some hash code of the entity object. There is no guaranteed pattern for it.
Thanks for the explanation! Can you provide the corresponding tables for those symbols and the entity object? By the way, is there a list of functions such as "toHour", etc provided?
I noticed that in the omnt text data created, there are entity@ in the input tokens. E.g.
But it seems there is no explanation for entity@. Could you provide an explanation for that?