my-project-public

repository

loading code, commits, and activity

repositories

loading repo index

#1	---
#2	title: Azure AI Search
#3	---
#4
#5	[Azure AI Search](https://learn.microsoft.com/azure/search/search-what-is-azure-search/) (formerly known as "Azure Cognitive Search") provides secure information retrieval at scale over user-owned content in traditional and generative AI search applications.
#6
#7	## Usage
#8
#9	```python
#10	import os
#11	from mem0 import Memory
#12
#13	os.environ["OPENAI_API_KEY"] = "sk-xx" # This key is used for embedding purpose
#14
#15	config = {
#16	"vector_store": {
#17	"provider": "azure_ai_search",
#18	"config": {
#19	"service_name": "<your-azure-ai-search-service-name>",
#20	"api_key": "<your-api-key>",
#21	"collection_name": "mem0",
#22	"embedding_model_dims": 1536
#23	}
#24	}
#25	}
#26
#27	m = Memory.from_config(config)
#28	messages = [
#29	{"role": "user", "content": "I'm planning to watch a movie tonight. Any recommendations?"},
#30	{"role": "assistant", "content": "How about thriller movies? They can be quite engaging."},
#31	{"role": "user", "content": "I'm not a big fan of thriller movies but I love sci-fi movies."},
#32	{"role": "assistant", "content": "Got it! I'll avoid thriller recommendations and suggest sci-fi movies in the future."}
#33	]
#34	m.add(messages, user_id="alice", metadata={"category": "movies"})
#35	```
#36
#37	## Using binary compression for large vector collections
#38
#39	```python
#40	config = {
#41	"vector_store": {
#42	"provider": "azure_ai_search",
#43	"config": {
#44	"service_name": "<your-azure-ai-search-service-name>",
#45	"api_key": "<your-api-key>",
#46	"collection_name": "mem0",
#47	"embedding_model_dims": 1536,
#48	"compression_type": "binary",
#49	"use_float16": True # Use half precision for storage efficiency
#50	}
#51	}
#52	}
#53	```
#54
#55	## Using hybrid search
#56
#57	```python
#58	config = {
#59	"vector_store": {
#60	"provider": "azure_ai_search",
#61	"config": {
#62	"service_name": "<your-azure-ai-search-service-name>",
#63	"api_key": "<your-api-key>",
#64	"collection_name": "mem0",
#65	"embedding_model_dims": 1536,
#66	"hybrid_search": True,
#67	"vector_filter_mode": "postFilter"
#68	}
#69	}
#70	}
#71	```
#72
#73	## Using Azure Identity for Authentication
#74	As an alternative to using an API key, the Azure Identity credential chain can be used to authenticate with Azure OpenAI. The list below shows the order of precedence for credential application:
#75
#76	1. Environment Credential:
#77	Azure client ID, secret, tenant ID, or certificate in environment variables for service principal authentication.
#78
#79	2. Workload Identity Credential:
#80	Utilizes Azure Workload Identity (relevant for Kubernetes and Azure workloads).
#81
#82	3. Managed Identity Credential:
#83	Authenticates as a Managed Identity (for apps/services hosted in Azure with Managed Identity enabled), this is the most secure production credential.
#84
#85	4. Shared Token Cache Credential / Visual Studio Credential (Windows only):
#86	Uses cached credentials from Visual Studio sign-ins (and sometimes VS Code if SSO is enabled).
#87
#88	5. Azure CLI Credential:
#89	Uses the currently logged-in user from the Azure CLI (`az login`), this is the most common development credential.
#90
#91	6. Azure PowerShell Credential:
#92	Uses the identity from Azure PowerShell (`Connect-AzAccount`).
#93
#94	7. Azure Developer CLI Credential:
#95	Uses the session from Azure Developer CLI (`azd auth login`).
#96
#97	<Note> If an API is provided, it will be used for authentication over an Azure Identity </Note>
#98	To enable Role-Based Access Control (RBAC) for Azure AI Search, follow these steps:
#99
#100	1. In the Azure Portal, navigate to your Azure AI Search service.
#101	2. In the left menu, select Settings > Keys.
#102	3. Change the authentication setting to Role-based access control, or Both if you need API key compatibility. The default is “Key-based authentication”—you must switch it to use Azure roles.
#103	4. Go to Access Control (IAM):
#104	- In the Azure Portal, select your Search service.
#105	- Click Access Control (IAM) on the left.
#106	5. Add a Role Assignment:
#107	- Click Add > Add role assignment.
#108	6. Choose Role:
#109	- Mem0 requires the Search Index Data Contributor and Search Service Contributor role.
#110	7. Choose Member
#111	- To assign to a User, Group, Service Principal or Managed Identity:
#112	- For production it is recommended to use a service principal or managed identity.
#113	- For a service principal: select User, group, or service principal and search for the service principal.
#114	- For a managed identity: select Managed identity and choose the managed identity.
#115	- For development, you can assign the role to a user account.
#116	- For development: select User, group, or service principal and pick an Azure Entra ID account (the same used with `az login`).
#117	8. Complete the Assignment:
#118	- Click Review + Assign.
#119
#120	If you are using Azure Identity, do not set the `api_key` in the configuration.
#121	```python
#122	config = {
#123	"vector_store": {
#124	"provider": "azure_ai_search",
#125	"config": {
#126	"service_name": "<your-azure-ai-search-service-name>",
#127	"collection_name": "mem0",
#128	"embedding_model_dims": 1536,
#129	"compression_type": "binary",
#130	"use_float16": True # Use half precision for storage efficiency
#131	}
#132	}
#133	}
#134	```
#135
#136	### Environment Variables to Use Azure Identity Credential
#137	* For an Environment Credential, you will need to setup a Service Principal and set the following environment variables:
#138	- `AZURE_TENANT_ID`: Your Azure Active Directory tenant ID.
#139	- `AZURE_CLIENT_ID`: The client ID of your service principal or managed identity.
#140	- `AZURE_CLIENT_SECRET`: The client secret of your service principal.
#141	* For a User-Assigned Managed Identity, you will need to set the following environment variable:
#142	- `AZURE_CLIENT_ID`: The client ID of the user-assigned managed identity.
#143	* For a System-Assigned Managed Identity, no additional environment variables are needed.
#144
#145	### Developer Logins for Azure Identity Credential
#146	* For an Azure CLI Credential, you need to have the Azure CLI installed and logged in with `az login`.
#147	* For an Azure PowerShell Credential, you need to have the Azure PowerShell module installed and logged in with `Connect-AzAccount`.
#148	* For an Azure Developer CLI Credential, you need to have the Azure Developer CLI installed and logged in with `azd auth login`.
#149
#150	Troubleshooting tips for [Azure Identity](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/identity/azure-identity/TROUBLESHOOTING.md#troubleshoot-environmentcredential-authentication-issues).
#151
#152
#153	## Configuration Parameters
#154
#155	\| Parameter \| Description \| Default Value \| Options \|
#156	\| --- \| --- \| --- \| --- \|
#157	\| `service_name` \| Azure AI Search service name \| Required \| - \|
#158	\| `api_key` \| API key of the Azure AI Search service \| Optional \| If not present, the [Azure Identity](#using-azure-identity-for-authentication) credential chain will be used \|
#159	\| `collection_name` \| The name of the collection/index to store vectors \| `mem0` \| Any valid index name \|
#160	\| `embedding_model_dims` \| Dimensions of the embedding model \| `1536` \| Any integer value \|
#161	\| `compression_type` \| Type of vector compression to use \| `none` \| `none`, `scalar`, `binary` \|
#162	\| `use_float16` \| Store vectors in half precision (Edm.Half) \| `False` \| `True`, `False` \|
#163	\| `vector_filter_mode` \| Vector filter mode to use \| `preFilter` \| `postFilter`, `preFilter` \|
#164	\| `hybrid_search` \| Use hybrid search \| `False` \| `True`, `False` \|
#165
#166	## Notes on Configuration Options
#167
#168	- compression_type:
#169	- `none`: No compression, uses full vector precision
#170	- `scalar`: Scalar quantization with reasonable balance of speed and accuracy
#171	- `binary`: Binary quantization for maximum compression with some accuracy trade-off
#172
#173	- vector_filter_mode:
#174	- `preFilter`: Applies filters before vector search (faster)
#175	- `postFilter`: Applies filters after vector search (may provide better relevance)
#176
#177	- use_float16: Using half precision (float16) reduces storage requirements but may slightly impact accuracy. Useful for very large vector collections.
#178
#179	- Filterable Fields: The implementation automatically extracts `user_id`, `run_id`, and `agent_id` fields from payloads for filtering.

z6Mkq5mY3JWtxoxUobWcfNHm7AkRubgSWEZTkBVqZXJviFZ5/my-project-public