caikit.interfaces.common.data_model.vectors

Data structures for embedding vector representations

Attributes

`log`
`error`

`PyFloatSequence`	A DataObject is a data model class that is backed by a @dataclass.
`NpFloat32Sequence`	A DataObject is a data model class that is backed by a @dataclass.
`NpFloat64Sequence`	A DataObject is a data model class that is backed by a @dataclass.
`Vector1D`	Data representation for a 1 dimension vector of float-type data.
`ListOfVector1D`	Data representation for an embedding matrix holding 2D vectors

class caikit.interfaces.common.data_model.vectors.PyFloatSequence[source]

A DataObject is a data model class that is backed by a @dataclass.

Data model classes that use the @dataobject decorator must derive from this base class.

values: py_to_proto.dataclass_to_proto.Annotated[List[float], FieldNumber(1)] = []

class caikit.interfaces.common.data_model.vectors.NpFloat32Sequence[source]

A DataObject is a data model class that is backed by a @dataclass.

Data model classes that use the @dataobject decorator must derive from this base class.

values: py_to_proto.dataclass_to_proto.Annotated[List[numpy.float32], FieldNumber(1)]

classmethod from_proto(proto)[source]

Build a DataBase from protobufs.

class caikit.interfaces.common.data_model.vectors.NpFloat64Sequence[source]

A DataObject is a data model class that is backed by a @dataclass.

Data model classes that use the @dataobject decorator must derive from this base class.

values: py_to_proto.dataclass_to_proto.Annotated[List[numpy.float64], FieldNumber(1)]

classmethod from_proto(proto)[source]

Build a DataBase from protobufs.

class caikit.interfaces.common.data_model.vectors.Vector1D[source]

Data representation for a 1 dimension vector of float-type data.

data: py_to_proto.dataclass_to_proto.Annotated[PyFloatSequence | NpFloat32Sequence | NpFloat64Sequence, FieldNumber(1)]

classmethod from_json(json_str: Dict[str, Any] | str) → Vector1D[source]: JSON does not have different float types. Move data into data_pyfloatsequence

classmethod from_proto(proto)[source]: Wrap the data in an appropriate float sequence, wrapped by this class

fill_proto(proto)[source]: Fill in the data in an appropriate data_<float type sequence>

class caikit.interfaces.common.data_model.vectors.ListOfVector1D[source]

Data representation for an embedding matrix holding 2D vectors

vectors: py_to_proto.dataclass_to_proto.Annotated[List[Vector1D], FieldNumber(1)]

classmethod from_json(json_str: Dict[str, Any] | str) → ListOfVector1D[source]: Fill in the vector data in an appropriate data_<float type sequence>