Skip to content

Fuzzy Alignment Threshold Setting not Working #245

@difersalest

Description

@difersalest

I was testing the extract function with a different fuzzy threshold and it seems it is not being unpacked in the resolver function:
Code:
result = lx.extract(
text_or_documents=input_transcript,
prompt_description=PROMPT_SOP,
examples=EXAMPLES_SOP,
model=model,
extraction_passes=1,
max_char_buffer=5000,
fetch_urls=False,
resolver_params={'fuzzy_alignment_threshold': 0.5},
)

Error:

Image

resolver.py constructor:

class Resolver(AbstractResolver):

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions