AlibabaResearch / AlibabaResearch/DAMO-ConvAI
Graphix inference code for Spider test set and dev/test of variants of Spider (DK, SYN, Realistic)
- Vorherrschende Sprache
- Python
- Sterne
- 1.6k
- Forks
- 250
- PR-Merge-Kennzahlen
- Keine gemergten PRs in 30 T.
Beschreibung
Hi there,
I've been trying to replicate the performance of the `graphix-3b + picard` model on the Spider test set, but I noticed that there doesn't seem to be an existing script for this. Could anyone please share the script/code-snippet used to do so?
Also, in order to replicate the results of the model on **spider-syn**, **spider-dk** and **spider-realistic**, it seems that we need to have the following python scripts which are currently missing in the repo: `seq2seq/datasets/spider-syn/spider-syn.py`, `seq2seq/datasets/spider-dk/spider-dk.py` and `seq2seq/datasets/spider-realistic/spider-realistic.py`
Thanks a ton for assisting with this!
Beitragsleitfaden
Für dieses Repository ist kein Beitragsleitfaden indexiert
Bewertung
Dieses Issue wurde noch nicht bewertet.