Skip to content

Dataframe loader

DataFrameLoader

Bases: BaseLoader

Source code in griptape/loaders/dataframe_loader.py 
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
@define
class DataFrameLoader(BaseLoader):
    embedding_driver: Optional[BaseEmbeddingDriver] = field(default=None, kw_only=True)

    def load(self, source: DataFrame, *args, **kwargs) -> list[CsvRowArtifact]:
        artifacts = []

        chunks = [CsvRowArtifact(row) for row in source.to_dict(orient="records")]

        if self.embedding_driver:
            for chunk in chunks:
                chunk.generate_embedding(self.embedding_driver)

        for chunk in chunks:
            artifacts.append(chunk)

        return artifacts

    def load_collection(self, sources: list[DataFrame], *args, **kwargs) -> dict[str, list[CsvRowArtifact]]:
        return cast(dict[str, list[CsvRowArtifact]], super().load_collection(sources, *args, **kwargs))

    def to_key(self, source: DataFrame, *args, **kwargs) -> str:
        hash_pandas_object = import_optional_dependency("pandas.core.util.hashing").hash_pandas_object

        return str_to_hash(str(hash_pandas_object(source, index=True).values))

embedding_driver: Optional[BaseEmbeddingDriver] = field(default=None, kw_only=True) class-attribute instance-attribute

load(source, *args, **kwargs)

Source code in griptape/loaders/dataframe_loader.py 
21
22
23
24
25
26
27
28
29
30
31
32
33
def load(self, source: DataFrame, *args, **kwargs) -> list[CsvRowArtifact]:
    artifacts = []

    chunks = [CsvRowArtifact(row) for row in source.to_dict(orient="records")]

    if self.embedding_driver:
        for chunk in chunks:
            chunk.generate_embedding(self.embedding_driver)

    for chunk in chunks:
        artifacts.append(chunk)

    return artifacts

load_collection(sources, *args, **kwargs)

Source code in griptape/loaders/dataframe_loader.py 
35
36
def load_collection(self, sources: list[DataFrame], *args, **kwargs) -> dict[str, list[CsvRowArtifact]]:
    return cast(dict[str, list[CsvRowArtifact]], super().load_collection(sources, *args, **kwargs))

to_key(source, *args, **kwargs)

Source code in griptape/loaders/dataframe_loader.py 
38
39
40
41
def to_key(self, source: DataFrame, *args, **kwargs) -> str:
    hash_pandas_object = import_optional_dependency("pandas.core.util.hashing").hash_pandas_object

    return str_to_hash(str(hash_pandas_object(source, index=True).values))