Repository files navigation

postgresml-django is a Python module that integrates PostgresML with Django ORM, enabling automatic in-database embedding of Django models. It simplifies the process of creating and searching vector embeddings for your text data.

This module provides a seamless way to:

• Automatically generate in-databse embeddings for specified fields in your Django models
• Perform vector similarity searches in-database

1. Ensure you havepgml installed and configured in your database. The easiest way to do that is to sign up for a free serverless database atpostgresml.org. You can also host it your self.

2. Install the package using pip:

   pip install postgresml-django

You are ready to go!

This example demonstrates using theintfloat/e5-small-v2 transformer, which has an embedding size of 384.

fromdjango.dbimportmodelsfrompostgresml_djangoimportVectorField,EmbedclassDocument(Embed):text=models.TextField()text_embedding=VectorField(field_to_embed="text",dimensions=384,transformer="intfloat/e5-small-v2"    )# Searchingresults=Document.vector_search("text_embedding","some query to search against")

This example shows how to use themixedbread-ai/mxbai-embed-large-v1 transformer, which has an embedding size of 1024 and requires specific parameters for recall.

fromdjango.dbimportmodelsfrompostgresml_djangoimportVectorField,EmbedclassArticle(Embed):content=models.TextField()content_embedding=VectorField(field_to_embed="content",dimensions=1024,transformer="mixedbread-ai/mxbai-embed-large-v1",transformer_recall_parameters={"prompt":"Represent this sentence for searching relevant passages: "        }    )# Searchingresults=Article.vector_search("content_embedding","some query to search against")

Note the differences between the two examples:

1. Thedimensions parameter is set to 384 forintfloat/e5-small-v2 and 1024 formixedbread-ai/mxbai-embed-large-v1.
2. Themixedbread-ai/mxbai-embed-large-v1 transformer requires additional parameters for recall, which are specified in thetransformer_recall_parameters argument.

Both examples will automatically generate embeddings when instances are saved and allow for vector similarity searches using thevector_search method.

We welcome contributions to postgresml-django! Whether it's bug reports, feature requests, documentation improvements, or code contributions, your input is valuable to us. Feel free to open issues or submit pull requests on our GitHub repository.