swarmauri_embedding_doc2vec 0.11.0.dev1

A Doc2Vec based Embedding Model.

Swarmauri Embedding Doc2vec

A Gensim-powered Doc2Vec implementation for document embeddings in the Swarmauri ecosystem. The component registers as Doc2VecEmbedding and returns vectors as swarmauri_standard.vectors.Vector instances.

Installation

Install the package with your preferred Python packaging tool:

pip install swarmauri_embedding_doc2vec

poetry add swarmauri_embedding_doc2vec

uv pip install swarmauri_embedding_doc2vec

Usage

from swarmauri_embedding_doc2vec import Doc2VecEmbedding

documents = [
    "This is the first document.",
    "Here is another document.",
    "And a third one.",
]

# Initialize the embedder. Adjust parameters to match your dataset size.
embedder = Doc2VecEmbedding(vector_size=300, window=10, min_count=1, workers=1)

# Fit and transform documents into Vector objects.
vectors = embedder.fit_transform(documents)

# Access the raw embedding values via the Vector.value attribute.
first_vector = vectors[0].value

# Transform new documents (the result is also a Vector).
new_vector = embedder.transform(["This is a new document."])[0]

# Save and load the underlying Doc2Vec model.
model_path = "doc2vec.model"
embedder.save_model(model_path)
embedder.load_model(model_path)

Want to help?

If you want to contribute to swarmauri-sdk, read up on our guidelines for contributing that will help you get started.