EmbeddedChunk#

class scikitplot.corpus.EmbeddedChunk(chunk, embedding, model_name, embedding_dim)[source]#

A Chunk decorated with a dense embedding vector.

Parameters:

chunkChunk

The original text chunk.

embeddingEmbeddingVector

Dense float vector produced by an encoder model.

model_namestr

Name / identifier of the encoder model used.

embedding_dimint

Dimensionality of the embedding vector.

Parameters:

• chunk (Chunk)

• embedding (list[float])

• model_name (str)

• embedding_dim (int)

Notes

embedding_dim is stored explicitly so consumers can validate without computing len(embedding) repeatedly.

chunk: Chunk[source]#

embedding: list[float][source]#

embedding_dim: int[source]#

model_name: str[source]#

validate_dimension()[source]#

Raise if len(embedding) != embedding_dim.

Raises:

ValueError

If the embedding length does not match embedding_dim.

Return type:

None