ovos-chromadb-embeddings-plugin 0.0.0__tar.gz

ovos-chromadb-embeddings-plugin-0.0.0/PKG-INFO

@@ -0,0 +1,67 @@
+Metadata-Version: 2.1
+Name: ovos-chromadb-embeddings-plugin
+Version: 0.0.0
+Summary: A chromadb embeddings plugin for OVOS
+Home-page: https://github.com/TigreGotico/ovos-chromadb-embeddings-plugin
+Author: jarbasai
+Author-email: jarbasai@mailfence.com
+License: MIT
+Description: # ChromaEmbeddingsDB Plugin
+        
+        ## Overview
+        
+        The `ChromaEmbeddingsDB` plugin integrates with the [ChromaDB](https://www.trychroma.com/) database to provide a robust solution for managing and querying embeddings. This plugin extends the abstract `EmbeddingsDB` class, allowing you to store, retrieve, and query embeddings efficiently using ChromaDB’s capabilities.
+        
+        This plugin is meant to be used by other specialized plugins such as:
+        - https://github.com/TigreGotico/ovos-face-embeddings-plugin
+        - https://github.com/TigreGotico/ovos-voice-embeddings-plugin
+        - https://github.com/TigreGotico/ovos-gguf-embeddings-plugin
+        
+        ## Features
+        
+        - **Add Embeddings**: Store embeddings with associated keys.
+        - **Retrieve Embeddings**: Fetch embeddings by their keys.
+        - **Delete Embeddings**: Remove embeddings by their keys.
+        - **Query Embeddings**: Find the closest embeddings to a given query, with support for cosine distance.
+        
+        ## Example
+        
+        Here is a full example demonstrating the basic usage of `ChromaEmbeddingsDB`.
+        
+        ```python
+        import numpy as np
+        from chroma_embeddings_db import ChromaEmbeddingsDB
+        
+        # Initialize the database
+        db = ChromaEmbeddingsDB(path="path_to_chromadb_storage")
+        
+        # Add embeddings
+        embedding1 = np.array([0.1, 0.2, 0.3])
+        embedding2 = np.array([0.4, 0.5, 0.6])
+        db.add_embeddings("user1", embedding1)
+        db.add_embeddings("user2", embedding2)
+        
+        # Retrieve and print embeddings
+        print(db.get_embedding("user1"))
+        print(db.get_embedding("user2"))
+        
+        # Query embeddings
+        query_embedding = np.array([0.2, 0.3, 0.4])
+        results = db.query(query_embedding, top_k=2)
+        print(results)
+        
+        # Delete an embedding
+        db.delete_embedding("user1")
+        ```
+        
+        > Ensure that the path provided to the `ChromaEmbeddingsDB` constructor is accessible and writable.
+        
+        
+        ## Acknowledgements
+        
+        - [ChromaDB](https://www.trychroma.com/) for providing the database backend.
+        - [NumPy](https://numpy.org) for numerical operations.
+        
+Keywords: OVOS openvoiceos plugin chromadb recognition
+Platform: UNKNOWN
+Description-Content-Type: text/markdown

ovos-chromadb-embeddings-plugin-0.0.0/README.md

@@ -0,0 +1,55 @@
+# ChromaEmbeddingsDB Plugin
+
+## Overview
+
+The `ChromaEmbeddingsDB` plugin integrates with the [ChromaDB](https://www.trychroma.com/) database to provide a robust solution for managing and querying embeddings. This plugin extends the abstract `EmbeddingsDB` class, allowing you to store, retrieve, and query embeddings efficiently using ChromaDB’s capabilities.
+
+This plugin is meant to be used by other specialized plugins such as:
+- https://github.com/TigreGotico/ovos-face-embeddings-plugin
+- https://github.com/TigreGotico/ovos-voice-embeddings-plugin
+- https://github.com/TigreGotico/ovos-gguf-embeddings-plugin
+
+## Features
+
+- **Add Embeddings**: Store embeddings with associated keys.
+- **Retrieve Embeddings**: Fetch embeddings by their keys.
+- **Delete Embeddings**: Remove embeddings by their keys.
+- **Query Embeddings**: Find the closest embeddings to a given query, with support for cosine distance.
+
+## Example
+
+Here is a full example demonstrating the basic usage of `ChromaEmbeddingsDB`.
+
+```python
+import numpy as np
+from chroma_embeddings_db import ChromaEmbeddingsDB
+
+# Initialize the database
+db = ChromaEmbeddingsDB(path="path_to_chromadb_storage")
+
+# Add embeddings
+embedding1 = np.array([0.1, 0.2, 0.3])
+embedding2 = np.array([0.4, 0.5, 0.6])
+db.add_embeddings("user1", embedding1)
+db.add_embeddings("user2", embedding2)
+
+# Retrieve and print embeddings
+print(db.get_embedding("user1"))
+print(db.get_embedding("user2"))
+
+# Query embeddings
+query_embedding = np.array([0.2, 0.3, 0.4])
+results = db.query(query_embedding, top_k=2)
+print(results)
+
+# Delete an embedding
+db.delete_embedding("user1")
+```
+
+> Ensure that the path provided to the `ChromaEmbeddingsDB` constructor is accessible and writable.
+
+
+## Acknowledgements
+
+- [ChromaDB](https://www.trychroma.com/) for providing the database backend.
+- [NumPy](https://numpy.org) for numerical operations.

ovos-chromadb-embeddings-plugin-0.0.0/ovos_chromadb_embeddings/__init__.py

@@ -0,0 +1,112 @@
+from typing import List, Optional, Dict
+
+import chromadb
+import numpy as np
+
+from ovos_plugin_manager.templates.embeddings import EmbeddingsDB, EmbeddingsTuple
+
+
+class ChromaEmbeddingsDB(EmbeddingsDB):
+    """An implementation of EmbeddingsDB using ChromaDB for managing embeddings."""
+
+    def __init__(self, path: str):
+        """Initialize the ChromaEmbeddingsDB.
+
+        Args:
+            path (str): The path to the ChromaDB storage.
+        """
+        super().__init__()
+        self.client = chromadb.PersistentClient(path=path)
+        self.collection = self.client.get_or_create_collection(
+            "embeddings", metadata={"hnsw:space": "cosine"}
+        )
+
+    def add_embeddings(self, key: str, embedding: np.ndarray, metadata: Optional[Dict[str, any]] = None) -> None:
+        """Add or update embeddings in the database.
+
+        Args:
+            key (str): The unique key for the embedding.
+            embedding (np.ndarray): The embedding vector to store.
+            metadata (Optional[Dict[str, any]]): Optional metadata associated with the embedding.
+        """
+        self.collection.upsert(
+            embeddings=[embedding.tolist()],
+            ids=[key],
+            metadatas=[metadata or {}]
+        )
+
+    def delete_embedding(self, key: str) -> None:
+        """Delete embeddings from the database.
+
+        Args:
+            key (str): The unique key for the embedding to delete.
+        """
+        self.collection.delete(ids=[key])
+
+    def get_embedding(self, key: str) -> np.ndarray:
+        """Retrieve embeddings from the database.
+
+        Args:
+            key (str): The unique key for the embedding to retrieve.
+
+        Returns:
+            np.ndarray: The retrieved embedding vector.
+        """
+        result = self.collection.get(ids=[key], include=["embeddings"])
+        embedding_list = result['embeddings'][0]
+        return np.array(embedding_list)
+
+    def query(self, embedding: np.ndarray, top_k: int = 5, return_metadata: bool = False) -> List[EmbeddingsTuple]:
+        """Query the database for the closest embeddings to the given query embedding.
+
+        Args:
+            embedding (np.ndarray): The embedding vector to query.
+            top_k (int, optional): The number of top results to return. Defaults to 5.
+            return_metadata (bool, optional): Whether to include metadata in the results. Defaults to False.
+
+        Returns:
+            List[EmbeddingsTuple]: A list of tuples containing the keys, distances, and optionally metadata.
+        """
+        embedding_list = embedding.tolist() if isinstance(embedding, np.ndarray) else embedding
+        results = self.collection.query(
+            query_embeddings=[embedding_list],
+            n_results=top_k
+        )
+        ids = results["ids"][0]
+        distances = results["distances"][0]
+        metadatas = results["metadatas"][0]
+
+        if return_metadata:
+            return list(zip(ids, distances, metadatas))
+        return list(zip(ids, distances))
+
+
+if __name__ == "__main__":
+    import numpy as np
+
+    # Initialize the database
+    db = ChromaEmbeddingsDB(path="chromadb_storage")
+
+    # Add embeddings
+    embedding1 = np.array([0.1, 0.2, 0.3])
+    embedding2 = np.array([0.4, 0.5, 0.6])
+    db.add_embeddings("user1", embedding1, metadata={"name": "Bob"})
+    db.add_embeddings("user2", embedding2, metadata={"name": "Joe"})
+
+    # Retrieve and print embeddings
+    print(db.get_embedding("user1"))
+    print(db.get_embedding("user2"))
+
+    # Query embeddings
+    query_embedding = np.array([0.2, 0.3, 0.4])
+    results = db.query(query_embedding, top_k=2)
+    print(results)
+    # [('user2', 0.0053884541273605535),
+    #  ('user1', 0.007416666029069874)]
+    results = db.query(query_embedding, top_k=2, return_metadata=True)
+    print(results)
+    # [('user2', 0.0053884541273605535, {'name': 'Joe'}),
+    #  ('user1', 0.007416666029069874, {'name': 'Bob'})]
+
+    # Delete an embedding
+    db.delete_embedding("user1")

ovos-chromadb-embeddings-plugin-0.0.0/ovos_chromadb_embeddings/version.py

@@ -0,0 +1,7 @@
+# The following lines are replaced during the release process.
+# START_VERSION_BLOCK
+VERSION_MAJOR = 0
+VERSION_MINOR = 0
+VERSION_BUILD = 0
+VERSION_ALPHA = 0
+# END_VERSION_BLOCK

ovos-chromadb-embeddings-plugin-0.0.0/ovos_chromadb_embeddings_plugin.egg-info/PKG-INFO

@@ -0,0 +1,67 @@
+Metadata-Version: 2.1
+Name: ovos-chromadb-embeddings-plugin
+Version: 0.0.0
+Summary: A chromadb embeddings plugin for OVOS
+Home-page: https://github.com/TigreGotico/ovos-chromadb-embeddings-plugin
+Author: jarbasai
+Author-email: jarbasai@mailfence.com
+License: MIT
+Description: # ChromaEmbeddingsDB Plugin
+        
+        ## Overview
+        
+        The `ChromaEmbeddingsDB` plugin integrates with the [ChromaDB](https://www.trychroma.com/) database to provide a robust solution for managing and querying embeddings. This plugin extends the abstract `EmbeddingsDB` class, allowing you to store, retrieve, and query embeddings efficiently using ChromaDB’s capabilities.
+        
+        This plugin is meant to be used by other specialized plugins such as:
+        - https://github.com/TigreGotico/ovos-face-embeddings-plugin
+        - https://github.com/TigreGotico/ovos-voice-embeddings-plugin
+        - https://github.com/TigreGotico/ovos-gguf-embeddings-plugin
+        
+        ## Features
+        
+        - **Add Embeddings**: Store embeddings with associated keys.
+        - **Retrieve Embeddings**: Fetch embeddings by their keys.
+        - **Delete Embeddings**: Remove embeddings by their keys.
+        - **Query Embeddings**: Find the closest embeddings to a given query, with support for cosine distance.
+        
+        ## Example
+        
+        Here is a full example demonstrating the basic usage of `ChromaEmbeddingsDB`.
+        
+        ```python
+        import numpy as np
+        from chroma_embeddings_db import ChromaEmbeddingsDB
+        
+        # Initialize the database
+        db = ChromaEmbeddingsDB(path="path_to_chromadb_storage")
+        
+        # Add embeddings
+        embedding1 = np.array([0.1, 0.2, 0.3])
+        embedding2 = np.array([0.4, 0.5, 0.6])
+        db.add_embeddings("user1", embedding1)
+        db.add_embeddings("user2", embedding2)
+        
+        # Retrieve and print embeddings
+        print(db.get_embedding("user1"))
+        print(db.get_embedding("user2"))
+        
+        # Query embeddings
+        query_embedding = np.array([0.2, 0.3, 0.4])
+        results = db.query(query_embedding, top_k=2)
+        print(results)
+        
+        # Delete an embedding
+        db.delete_embedding("user1")
+        ```
+        
+        > Ensure that the path provided to the `ChromaEmbeddingsDB` constructor is accessible and writable.
+        
+        
+        ## Acknowledgements
+        
+        - [ChromaDB](https://www.trychroma.com/) for providing the database backend.
+        - [NumPy](https://numpy.org) for numerical operations.
+        
+Keywords: OVOS openvoiceos plugin chromadb recognition
+Platform: UNKNOWN
+Description-Content-Type: text/markdown

ovos-chromadb-embeddings-plugin-0.0.0/ovos_chromadb_embeddings_plugin.egg-info/SOURCES.txt

@@ -0,0 +1,11 @@
+README.md
+setup.py
+ovos_chromadb_embeddings/__init__.py
+ovos_chromadb_embeddings/version.py
+ovos_chromadb_embeddings_plugin.egg-info/PKG-INFO
+ovos_chromadb_embeddings_plugin.egg-info/SOURCES.txt
+ovos_chromadb_embeddings_plugin.egg-info/dependency_links.txt
+ovos_chromadb_embeddings_plugin.egg-info/entry_points.txt
+ovos_chromadb_embeddings_plugin.egg-info/requires.txt
+ovos_chromadb_embeddings_plugin.egg-info/top_level.txt
+ovos_chromadb_embeddings_plugin.egg-info/zip-safe

ovos-chromadb-embeddings-plugin-0.0.0/ovos_chromadb_embeddings_plugin.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

ovos-chromadb-embeddings-plugin-0.0.0/ovos_chromadb_embeddings_plugin.egg-info/entry_points.txt

@@ -0,0 +1,3 @@
+[opm.embeddings]
+ovos-chromadb-embeddings-plugin=ovos_chromadb_embeddings:ChromaEmbeddingsDB
+

ovos-chromadb-embeddings-plugin-0.0.0/ovos_chromadb_embeddings_plugin.egg-info/requires.txt

@@ -0,0 +1,3 @@
+ovos-plugin-manager
+numpy
+chromadb

ovos-chromadb-embeddings-plugin-0.0.0/ovos_chromadb_embeddings_plugin.egg-info/top_level.txt

@@ -0,0 +1 @@
+ovos_chromadb_embeddings

ovos-chromadb-embeddings-plugin-0.0.0/ovos_chromadb_embeddings_plugin.egg-info/zip-safe

@@ -0,0 +1 @@
+

ovos-chromadb-embeddings-plugin-0.0.0/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

ovos-chromadb-embeddings-plugin-0.0.0/setup.py

@@ -0,0 +1,65 @@
+#!/usr/bin/env python3
+import os
+
+from setuptools import setup
+
+BASEDIR = os.path.abspath(os.path.dirname(__file__))
+
+with open(f"{BASEDIR}/README.md", "r") as fh:
+    long_desc = fh.read()
+
+
+def get_version():
+    """ Find the version of the package"""
+    version_file = os.path.join(BASEDIR, 'ovos_chromadb_embeddings', 'version.py')
+    major, minor, build, alpha = (None, None, None, None)
+    with open(version_file) as f:
+        for line in f:
+            if 'VERSION_MAJOR' in line:
+                major = line.split('=')[1].strip()
+            elif 'VERSION_MINOR' in line:
+                minor = line.split('=')[1].strip()
+            elif 'VERSION_BUILD' in line:
+                build = line.split('=')[1].strip()
+            elif 'VERSION_ALPHA' in line:
+                alpha = line.split('=')[1].strip()
+
+            if ((major and minor and build and alpha) or
+                    '# END_VERSION_BLOCK' in line):
+                break
+    version = f"{major}.{minor}.{build}"
+    if int(alpha):
+        version += f"a{alpha}"
+    return version
+
+
+def required(requirements_file):
+    """ Read requirements file and remove comments and empty lines. """
+    with open(os.path.join(BASEDIR, requirements_file), 'r') as f:
+        requirements = f.read().splitlines()
+        if 'MYCROFT_LOOSE_REQUIREMENTS' in os.environ:
+            print('USING LOOSE REQUIREMENTS!')
+            requirements = [r.replace('==', '>=').replace('~=', '>=') for r in requirements]
+        return [pkg for pkg in requirements
+                if pkg.strip() and not pkg.startswith("#")]
+
+
+PLUGIN_ENTRY_POINT = 'ovos-chromadb-embeddings-plugin=ovos_chromadb_embeddings:ChromaEmbeddingsDB'
+
+
+setup(
+    name='ovos-chromadb-embeddings-plugin',
+    version=get_version(),
+    description='A chromadb embeddings plugin for OVOS',
+    url='https://github.com/TigreGotico/ovos-chromadb-embeddings-plugin',
+    author='jarbasai',
+    author_email='jarbasai@mailfence.com',
+    license='MIT',
+    packages=['ovos_chromadb_embeddings'],
+    zip_safe=True,
+    keywords='OVOS openvoiceos plugin chromadb recognition',
+    entry_points={'opm.embeddings': PLUGIN_ENTRY_POINT},
+    install_requires=required("requirements.txt"),
+    long_description=long_desc,
+    long_description_content_type='text/markdown'
+)