mnemosine.node.Node.add_embedding

Kind: method

Signature

Node.add_embedding(self, text=None)

Documentation

Embed this node and store the vector.

Requires storage.embed_fn to be set. When text is omitted the node's content is used.

Args:

• text (str | None): Text to embed. Defaults to self.content.

Returns:

• list of float: The computed embedding vector.

Raises:

• EmbeddingRequired: If no embed_fn is configured or there is
• no text to embed.
• VectorError: If the embedding function returns an invalid vector.

Example:

>>> db.embed_fn = lambda t: [1.0, 0.0]  # any embedder
>>> doc.add_embedding()
[1.0, 0.0]