SCAI-BIO / datastew

Python library for intelligent data stewardship using Large Language Model (LLM) embeddings
https://pypi.org/project/datastew/
Apache License 2.0
4 stars 0 forks source link

Create a new dto encapsulating mapping and similaritites and refactor functions #20

Open tiadams opened 2 months ago

tiadams commented 2 months ago

function signature should be as follows:


class MappingVectorMatch:

     mapping: Mapping
     similarity: Float
def get_closest_mappings(self, embedding_model_name, terminology_name=None, embedding, limit=5) -> List[MappingVectorMatch]:
    mapping_matches = []
    if terminology is None:
    # per default search over all terminologies
    else:
    # do terminology specific search
    return mapping_matches
tiadams commented 1 month ago

terminology_name should support types:

tiadams commented 1 month ago

Makes sense to solve #21 before this to already use the new model class