Closed jeffreyrubi closed 2 months ago
In hybrid/vector search case, when calling client.search, get_embedding() is returning List[VectorQueries]: https://github.com/stanfordnlp/dspy/blob/main/dspy/retrieve/azureaisearch_rm.py#L380
It can directly passed to the client.search.
Currently, an extra '[]' is added when passing the vector_query to the client.search: https://github.com/stanfordnlp/dspy/blob/main/dspy/retrieve/azureaisearch_rm.py#L258
Results in serialization error from AISearch library.
The fix is to remove that extra []. Test is done with positive result after fix.
Thanks @jeffreyrubi !
In hybrid/vector search case, when calling client.search, get_embedding() is returning List[VectorQueries]: https://github.com/stanfordnlp/dspy/blob/main/dspy/retrieve/azureaisearch_rm.py#L380
It can directly passed to the client.search.
Currently, an extra '[]' is added when passing the vector_query to the client.search: https://github.com/stanfordnlp/dspy/blob/main/dspy/retrieve/azureaisearch_rm.py#L258
Results in serialization error from AISearch library.
The fix is to remove that extra []. Test is done with positive result after fix.