my-project-public

repository

loading code, commits, and activity

repositories

loading repo index

#1	[Elasticsearch](https://www.elastic.co/) is a distributed, RESTful search and analytics engine that can efficiently store and search vector data using dense vectors and k-NN search.
#2
#3	### Installation
#4
#5	Elasticsearch support requires additional dependencies. Install them with:
#6
#7	```bash
#8	pip install elasticsearch>=8.0.0
#9	```
#10
#11	### Usage
#12
#13	```python
#14	import os
#15	from mem0 import Memory
#16
#17	os.environ["OPENAI_API_KEY"] = "sk-xx"
#18
#19	config = {
#20	"vector_store": {
#21	"provider": "elasticsearch",
#22	"config": {
#23	"collection_name": "mem0",
#24	"host": "localhost",
#25	"port": 9200,
#26	"embedding_model_dims": 1536
#27	}
#28	}
#29	}
#30
#31	m = Memory.from_config(config)
#32	messages = [
#33	{"role": "user", "content": "I'm planning to watch a movie tonight. Any recommendations?"},
#34	{"role": "assistant", "content": "How about thriller movies? They can be quite engaging."},
#35	{"role": "user", "content": "I’m not a big fan of thriller movies but I love sci-fi movies."},
#36	{"role": "assistant", "content": "Got it! I'll avoid thriller recommendations and suggest sci-fi movies in the future."}
#37	]
#38	m.add(messages, user_id="alice", metadata={"category": "movies"})
#39	```
#40
#41	### Config
#42
#43	Here are the parameters available for configuring Elasticsearch:
#44
#45	\| Parameter \| Description \| Default Value \|
#46	\| ---------------------- \| -------------------------------------------------- \| ------------- \|
#47	\| `collection_name` \| The name of the index to store the vectors \| `mem0` \|
#48	\| `embedding_model_dims` \| Dimensions of the embedding model \| `1536` \|
#49	\| `host` \| The host where the Elasticsearch server is running \| `localhost` \|
#50	\| `port` \| The port where the Elasticsearch server is running \| `9200` \|
#51	\| `cloud_id` \| Cloud ID for Elastic Cloud deployment \| `None` \|
#52	\| `api_key` \| API key for authentication \| `None` \|
#53	\| `user` \| Username for basic authentication \| `None` \|
#54	\| `password` \| Password for basic authentication \| `None` \|
#55	\| `verify_certs` \| Whether to verify SSL certificates \| `True` \|
#56	\| `auto_create_index` \| Whether to automatically create the index \| `True` \|
#57	\| `custom_search_query` \| Function returning a custom search query \| `None` \|
#58	\| `headers` \| Custom headers to include in requests \| `None` \|
#59
#60	### Features
#61
#62	- Efficient vector search using Elasticsearch's native k-NN search
#63	- Support for both local and cloud deployments (Elastic Cloud)
#64	- Multiple authentication methods (Basic Auth, API Key)
#65	- Automatic index creation with optimized mappings for vector search
#66	- Memory isolation through payload filtering
#67	- Custom search query function to customize the search query
#68
#69	### Custom Search Query
#70
#71	The `custom_search_query` parameter allows you to customize the search query when `Memory.search` is called.
#72
#73	__Example__
#74	```python
#75	import os
#76	from typing import List, Optional, Dict
#77	from mem0 import Memory
#78
#79	def custom_search_query(query: List[float], limit: int, filters: Optional[Dict]) -> Dict:
#80	return {
#81	"knn": {
#82	"field": "vector",
#83	"query_vector": query,
#84	"k": limit,
#85	"num_candidates": limit * 2
#86	}
#87	}
#88
#89	os.environ["OPENAI_API_KEY"] = "sk-xx"
#90
#91	config = {
#92	"vector_store": {
#93	"provider": "elasticsearch",
#94	"config": {
#95	"collection_name": "mem0",
#96	"host": "localhost",
#97	"port": 9200,
#98	"embedding_model_dims": 1536,
#99	"custom_search_query": custom_search_query
#100	}
#101	}
#102	}
#103	```
#104	It should be a function that takes the following parameters:
#105	- `query`: a query vector used in `Memory.search`
#106	- `limit`: a number of results used in `Memory.search`
#107	- `filters`: a dictionary of key-value pairs used in `Memory.search`. You can add custom pairs for the custom search query.
#108
#109	The function should return a query body for the Elasticsearch search API.

z6Mkq5mY3JWtxoxUobWcfNHm7AkRubgSWEZTkBVqZXJviFZ5/my-project-public