Skip to main content

Redis

Redis vector database introduction and langchain integration guide.

What is Redis?โ€‹

Most developers from a web services background are familiar with Redis. At its core, Redis is an open-source key-value store that is used as a cache, message broker, and database. Developers choose Redis because it is fast, has a large ecosystem of client libraries, and has been deployed by major enterprises for years.

On top of these traditional use cases, Redis provides additional capabilities like the Search and Query capability that allows users to create secondary index structures within Redis. This allows Redis to be a Vector Database, at the speed of a cache.

Redis as a Vector Databaseโ€‹

Redis uses compressed, inverted indexes for fast indexing with a low memory footprint. It also supports a number of advanced features such as:

  • Indexing of multiple fields in Redis hashes and JSON
  • Vector similarity search (with HNSW (ANN) or FLAT (KNN))
  • Vector Range Search (e.g.ย find all vectors within a radius of a query vector)
  • Incremental indexing without performance loss
  • Document ranking (using tf-idf, with optional user-provided weights)
  • Field weighting
  • Complex boolean queries with AND, OR, and NOT operators
  • Prefix matching, fuzzy matching, and exact-phrase queries
  • Support for double-metaphone phonetic matching
  • Auto-complete suggestions (with fuzzy prefix suggestions)
  • Stemming-based query expansion in many languages (using Snowball)
  • Support for Chinese-language tokenization and querying (using Friso)
  • Numeric filters and ranges
  • Geospatial searches using Redis geospatial indexing
  • A powerful aggregations engine
  • Supports for all utf-8 encoded text
  • Retrieve full documents, selected fields, or only the document IDs
  • Sorting results (for example, by creation date)

Clientsโ€‹

Since Redis is much more than just a vector database, there are often use cases that demand the usage of a Redis client besides just the LangChain integration. You can use any standard Redis client library to run Search and Query commands, but itโ€™s easiest to use a library that wraps the Search and Query API. Below are a few examples, but you can find more client libraries here.

ProjectLanguageLicenseAuthorStars
jedisJavaMITRedisStars
redisvlPythonMITRedisStars
redis-pyPythonMITRedisStars
node-redisNode.jsMITRedisStars
nredisstack.NETMITRedisStars

Deployment optionsโ€‹

There are many ways to deploy Redis with RediSearch. The easiest way to get started is to use Docker, but there are are many potential options for deployment such as

Additional examplesโ€‹

Many examples can be found in the Redis AI teamโ€™s GitHub

More resourcesโ€‹

For more information on how to use Redis as a vector database, check out the following resources:

Setting upโ€‹

Install Redis Python clientโ€‹

Redis-py is the officially supported client by Redis. Recently released is the RedisVL client which is purpose-built for the Vector Database use cases. Both can be installed with pip.

%pip install --upgrade --quiet  redis redisvl langchain-openai tiktoken

We want to use OpenAIEmbeddings so we have to get the OpenAI API Key.

import getpass
import os

os.environ["OPENAI_API_KEY"] = getpass.getpass("OpenAI API Key:")
from langchain_openai import OpenAIEmbeddings

embeddings = OpenAIEmbeddings()

Deploy Redis locallyโ€‹

To locally deploy Redis, run:

docker run -d -p 6379:6379 -p 8001:8001 redis/redis-stack:latest

If things are running correctly you should see a nice Redis UI at http://localhost:8001. See the Deployment options section above for other ways to deploy.

Redis connection Url schemasโ€‹

Valid Redis Url schemas are: 1. redis:// - Connection to Redis standalone, unencrypted 2. rediss:// - Connection to Redis standalone, with TLS encryption 3. redis+sentinel:// - Connection to Redis server via Redis Sentinel, unencrypted 4. rediss+sentinel:// - Connection to Redis server via Redis Sentinel, booth connections with TLS encryption

More information about additional connection parameters can be found in the redis-py documentation.

# connection to redis standalone at localhost, db 0, no password
redis_url = "redis://localhost:6379"
# connection to host "redis" port 7379 with db 2 and password "secret" (old style authentication scheme without username / pre 6.x)
redis_url = "redis://:secret@redis:7379/2"
# connection to host redis on default port with user "joe", pass "secret" using redis version 6+ ACLs
redis_url = "redis://joe:secret@redis/0"

# connection to sentinel at localhost with default group mymaster and db 0, no password
redis_url = "redis+sentinel://localhost:26379"
# connection to sentinel at host redis with default port 26379 and user "joe" with password "secret" with default group mymaster and db 0
redis_url = "redis+sentinel://joe:secret@redis"
# connection to sentinel, no auth with sentinel monitoring group "zone-1" and database 2
redis_url = "redis+sentinel://redis:26379/zone-1/2"

# connection to redis standalone at localhost, db 0, no password but with TLS support
redis_url = "rediss://localhost:6379"
# connection to redis sentinel at localhost and default port, db 0, no password
# but with TLS support for booth Sentinel and Redis server
redis_url = "rediss+sentinel://localhost"

Sample dataโ€‹

First we will describe some sample data so that the various attributes of the Redis vector store can be demonstrated.

metadata = [
{
"user": "john",
"age": 18,
"job": "engineer",
"credit_score": "high",
},
{
"user": "derrick",
"age": 45,
"job": "doctor",
"credit_score": "low",
},
{
"user": "nancy",
"age": 94,
"job": "doctor",
"credit_score": "high",
},
{
"user": "tyler",
"age": 100,
"job": "engineer",
"credit_score": "high",
},
{
"user": "joe",
"age": 35,
"job": "dentist",
"credit_score": "medium",
},
]
texts = ["foo", "foo", "foo", "bar", "bar"]

Create Redis vector storeโ€‹

The Redis VectorStore instance can be initialized in a number of ways. There are multiple class methods that can be used to initialize a Redis VectorStore instance.

  • Redis.__init__ - Initialize directly
  • Redis.from_documents - Initialize from a list of Langchain.docstore.Document objects
  • Redis.from_texts - Initialize from a list of texts (optionally with metadata)
  • Redis.from_texts_return_keys - Initialize from a list of texts (optionally with metadata) and return the keys
  • Redis.from_existing_index - Initialize from an existing Redis index

Below we will use the Redis.from_texts method.

from langchain_community.vectorstores.redis import Redis

rds = Redis.from_texts(
texts,
embeddings,
metadatas=metadata,
redis_url="redis://localhost:6379",
index_name="users",
)
rds.index_name
'users'

Inspecting the created Indexโ€‹

Once the Redis VectorStore object has been constructed, an index will have been created in Redis if it did not already exist. The index can be inspected with both the rvland the redis-cli command line tool. If you installed redisvl above, you can use the rvl command line tool to inspect the index.

# assumes you're running Redis locally (use --host, --port, --password, --username, to change this)
!rvl index listall
16:58:26 [RedisVL] INFO   Indices:
16:58:26 [RedisVL] INFO 1. users

The Redis VectorStore implementation will attempt to generate index schema (fields for filtering) for any metadata passed through the from_texts, from_texts_return_keys, and from_documents methods. This way, whatever metadata is passed will be indexed into the Redis search index allowing for filtering on those fields.

Below we show what fields were created from the metadata we defined above

!rvl index info -i users


Index Information:
โ•ญโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ•ฎ
โ”‚ Index Name โ”‚ Storage Type โ”‚ Prefixes โ”‚ Index Options โ”‚ Indexing โ”‚
โ”œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ผโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ผโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ผโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ผโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ค
โ”‚ users โ”‚ HASH โ”‚ ['doc:users'] โ”‚ [] โ”‚ 0 โ”‚
โ•ฐโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ•ฏ
Index Fields:
โ•ญโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ•ฎ
โ”‚ Name โ”‚ Attribute โ”‚ Type โ”‚ Field Option โ”‚ Option Value โ”‚
โ”œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ผโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ผโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ผโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ผโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ค
โ”‚ user โ”‚ user โ”‚ TEXT โ”‚ WEIGHT โ”‚ 1 โ”‚
โ”‚ job โ”‚ job โ”‚ TEXT โ”‚ WEIGHT โ”‚ 1 โ”‚
โ”‚ credit_score โ”‚ credit_score โ”‚ TEXT โ”‚ WEIGHT โ”‚ 1 โ”‚
โ”‚ content โ”‚ content โ”‚ TEXT โ”‚ WEIGHT โ”‚ 1 โ”‚
โ”‚ age โ”‚ age โ”‚ NUMERIC โ”‚ โ”‚ โ”‚
โ”‚ content_vector โ”‚ content_vector โ”‚ VECTOR โ”‚ โ”‚ โ”‚
โ•ฐโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ•ฏ
!rvl stats -i users

Statistics:
โ•ญโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ฌโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ•ฎ
โ”‚ Stat Key โ”‚ Value โ”‚
โ”œโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ผโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ค
โ”‚ num_docs โ”‚ 5 โ”‚
โ”‚ num_terms โ”‚ 15 โ”‚
โ”‚ max_doc_id โ”‚ 5 โ”‚
โ”‚ num_records โ”‚ 33 โ”‚
โ”‚ percent_indexed โ”‚ 1 โ”‚
โ”‚ hash_indexing_failures โ”‚ 0 โ”‚
โ”‚ number_of_uses โ”‚ 4 โ”‚
โ”‚ bytes_per_record_avg โ”‚ 4.60606 โ”‚
โ”‚ doc_table_size_mb โ”‚ 0.000524521 โ”‚
โ”‚ inverted_sz_mb โ”‚ 0.000144958 โ”‚
โ”‚ key_table_size_mb โ”‚ 0.000193596 โ”‚
โ”‚ offset_bits_per_record_avg โ”‚ 8 โ”‚
โ”‚ offset_vectors_sz_mb โ”‚ 2.19345e-05 โ”‚
โ”‚ offsets_per_term_avg โ”‚ 0.69697 โ”‚
โ”‚ records_per_doc_avg โ”‚ 6.6 โ”‚
โ”‚ sortable_values_size_mb โ”‚ 0 โ”‚
โ”‚ total_indexing_time โ”‚ 0.32 โ”‚
โ”‚ total_inverted_index_blocks โ”‚ 16 โ”‚
โ”‚ vector_index_sz_mb โ”‚ 6.0126 โ”‚
โ•ฐโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”ดโ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ”€โ•ฏ

Itโ€™s important to note that we have not specified that the user, job, credit_score and age in the metadata should be fields within the index, this is because the Redis VectorStore object automatically generate the index schema from the passed metadata. For more information on the generation of index fields, see the API documentation.

Queryingโ€‹

There are multiple ways to query the Redis VectorStore implementation based on what use case you have:

  • similarity_search: Find the most similar vectors to a given vector.
  • similarity_search_with_score: Find the most similar vectors to a given vector and return the vector distance
  • similarity_search_limit_score: Find the most similar vectors to a given vector and limit the number of results to the score_threshold
  • similarity_search_with_relevance_scores: Find the most similar vectors to a given vector and return the vector similarities
  • max_marginal_relevance_search: Find the most similar vectors to a given vector while also optimizing for diversity
results = rds.similarity_search("foo")
print(results[0].page_content)
foo
# return metadata
results = rds.similarity_search("foo", k=3)
meta = results[1].metadata
print("Key of the document in Redis: ", meta.pop("id"))
print("Metadata of the document: ", meta)
Key of the document in Redis:  doc:users:a70ca43b3a4e4168bae57c78753a200f
Metadata of the document: {'user': 'derrick', 'job': 'doctor', 'credit_score': 'low', 'age': '45'}
# with scores (distances)
results = rds.similarity_search_with_score("foo", k=5)
for result in results:
print(f"Content: {result[0].page_content} --- Score: {result[1]}")
Content: foo --- Score: 0.0
Content: foo --- Score: 0.0
Content: foo --- Score: 0.0
Content: bar --- Score: 0.1566
Content: bar --- Score: 0.1566
# limit the vector distance that can be returned
results = rds.similarity_search_with_score("foo", k=5, distance_threshold=0.1)
for result in results:
print(f"Content: {result[0].page_content} --- Score: {result[1]}")
Content: foo --- Score: 0.0
Content: foo --- Score: 0.0
Content: foo --- Score: 0.0
# with scores
results = rds.similarity_search_with_relevance_scores("foo", k=5)
for result in results:
print(f"Content: {result[0].page_content} --- Similiarity: {result[1]}")
Content: foo --- Similiarity: 1.0
Content: foo --- Similiarity: 1.0
Content: foo --- Similiarity: 1.0
Content: bar --- Similiarity: 0.8434
Content: bar --- Similiarity: 0.8434
# limit scores (similarities have to be over .9)
results = rds.similarity_search_with_relevance_scores("foo", k=5, score_threshold=0.9)
for result in results:
print(f"Content: {result[0].page_content} --- Similarity: {result[1]}")
Content: foo --- Similarity: 1.0
Content: foo --- Similarity: 1.0
Content: foo --- Similarity: 1.0
# you can also add new documents as follows
new_document = ["baz"]
new_metadata = [{"user": "sam", "age": 50, "job": "janitor", "credit_score": "high"}]
# both the document and metadata must be lists
rds.add_texts(new_document, new_metadata)
['doc:users:b9c71d62a0a34241a37950b448dafd38']
# now query the new document
results = rds.similarity_search("baz", k=3)
print(results[0].metadata)
{'id': 'doc:users:b9c71d62a0a34241a37950b448dafd38', 'user': 'sam', 'job': 'janitor', 'credit_score': 'high', 'age': '50'}
# use maximal marginal relevance search to diversify results
results = rds.max_marginal_relevance_search("foo")
# the lambda_mult parameter controls the diversity of the results, the lower the more diverse
results = rds.max_marginal_relevance_search("foo", lambda_mult=0.1)

Connect to an existing Indexโ€‹

In order to have the same metadata indexed when using the Redis VectorStore. You will need to have the same index_schema passed in either as a path to a yaml file or as a dictionary. The following shows how to obtain the schema from an index and connect to an existing index.

# write the schema to a yaml file
rds.write_schema("redis_schema.yaml")

The schema file for this example should look something like:

numeric:
- name: age
no_index: false
sortable: false
text:
- name: user
no_index: false
no_stem: false
sortable: false
weight: 1
withsuffixtrie: false
- name: job
no_index: false
no_stem: false
sortable: false
weight: 1
withsuffixtrie: false
- name: credit_score
no_index: false
no_stem: false
sortable: false
weight: 1
withsuffixtrie: false
- name: content
no_index: false
no_stem: false
sortable: false
weight: 1
withsuffixtrie: false
vector:
- algorithm: FLAT
block_size: 1000
datatype: FLOAT32
dims: 1536
distance_metric: COSINE
initial_cap: 20000
name: content_vector

Notice, this include all possible fields for the schema. You can remove any fields that you donโ€™t need.

# now we can connect to our existing index as follows

new_rds = Redis.from_existing_index(
embeddings,
index_name="users",
redis_url="redis://localhost:6379",
schema="redis_schema.yaml",
)
results = new_rds.similarity_search("foo", k=3)
print(results[0].metadata)
{'id': 'doc:users:8484c48a032d4c4cbe3cc2ed6845fabb', 'user': 'john', 'job': 'engineer', 'credit_score': 'high', 'age': '18'}
# see the schemas are the same
new_rds.schema == rds.schema
True

Custom metadata indexingโ€‹

In some cases, you may want to control what fields the metadata maps to. For example, you may want the credit_score field to be a categorical field instead of a text field (which is the default behavior for all string fields). In this case, you can use the index_schema parameter in each of the initialization methods above to specify the schema for the index. Custom index schema can either be passed as a dictionary or as a path to a YAML file.

All arguments in the schema have defaults besides the name, so you can specify only the fields you want to change. All the names correspond to the snake/lowercase versions of the arguments you would use on the command line with redis-cli or in redis-py. For more on the arguments for each field, see the documentation

The below example shows how to specify the schema for the credit_score field as a Tag (categorical) field instead of a text field.

# index_schema.yml
tag:
- name: credit_score
text:
- name: user
- name: job
numeric:
- name: age

In Python, this would look like:


index_schema = {
"tag": [{"name": "credit_score"}],
"text": [{"name": "user"}, {"name": "job"}],
"numeric": [{"name": "age"}],
}

Notice that only the name field needs to be specified. All other fields have defaults.

# create a new index with the new schema defined above
index_schema = {
"tag": [{"name": "credit_score"}],
"text": [{"name": "user"}, {"name": "job"}],
"numeric": [{"name": "age"}],
}

rds, keys = Redis.from_texts_return_keys(
texts,
embeddings,
metadatas=metadata,
redis_url="redis://localhost:6379",
index_name="users_modified",
index_schema=index_schema, # pass in the new index schema
)
`index_schema` does not match generated metadata schema.
If you meant to manually override the schema, please ignore this message.
index_schema: {'tag': [{'name': 'credit_score'}], 'text': [{'name': 'user'}, {'name': 'job'}], 'numeric': [{'name': 'age'}]}
generated_schema: {'text': [{'name': 'user'}, {'name': 'job'}, {'name': 'credit_score'}], 'numeric': [{'name': 'age'}], 'tag': []}

The above warning is meant to notify users when they are overriding the default behavior. Ignore it if you are intentionally overriding the behavior.

Hybrid filteringโ€‹

With the Redis Filter Expression language built into LangChain, you can create arbitrarily long chains of hybrid filters that can be used to filter your search results. The expression language is derived from the RedisVL Expression Syntax and is designed to be easy to use and understand.

The following are the available filter types: - RedisText: Filter by full-text search against metadata fields. Supports exact, fuzzy, and wildcard matching. - RedisNum: Filter by numeric range against metadata fields. - RedisTag: Filter by the exact match against string-based categorical metadata fields. Multiple tags can be specified like โ€œtag1,tag2,tag3โ€.

The following are examples of utilizing these filters.


from langchain_community.vectorstores.redis import RedisText, RedisNum, RedisTag

# exact matching
has_high_credit = RedisTag("credit_score") == "high"
does_not_have_high_credit = RedisTag("credit_score") != "low"

# fuzzy matching
job_starts_with_eng = RedisText("job") % "eng*"
job_is_engineer = RedisText("job") == "engineer"
job_is_not_engineer = RedisText("job") != "engineer"

# numeric filtering
age_is_18 = RedisNum("age") == 18
age_is_not_18 = RedisNum("age") != 18
age_is_greater_than_18 = RedisNum("age") > 18
age_is_less_than_18 = RedisNum("age") < 18
age_is_greater_than_or_equal_to_18 = RedisNum("age") >= 18
age_is_less_than_or_equal_to_18 = RedisNum("age") <= 18

The RedisFilter class can be used to simplify the import of these filters as follows


from langchain_community.vectorstores.redis import RedisFilter

# same examples as above
has_high_credit = RedisFilter.tag("credit_score") == "high"
does_not_have_high_credit = RedisFilter.num("age") > 8
job_starts_with_eng = RedisFilter.text("job") % "eng*"

The following are examples of using a hybrid filter for search

from langchain_community.vectorstores.redis import RedisText

is_engineer = RedisText("job") == "engineer"
results = rds.similarity_search("foo", k=3, filter=is_engineer)

print("Job:", results[0].metadata["job"])
print("Engineers in the dataset:", len(results))
Job: engineer
Engineers in the dataset: 2
# fuzzy match
starts_with_doc = RedisText("job") % "doc*"
results = rds.similarity_search("foo", k=3, filter=starts_with_doc)

for result in results:
print("Job:", result.metadata["job"])
print("Jobs in dataset that start with 'doc':", len(results))
Job: doctor
Job: doctor
Jobs in dataset that start with 'doc': 2
from langchain_community.vectorstores.redis import RedisNum

is_over_18 = RedisNum("age") > 18
is_under_99 = RedisNum("age") < 99
age_range = is_over_18 & is_under_99
results = rds.similarity_search("foo", filter=age_range)

for result in results:
print("User:", result.metadata["user"], "is", result.metadata["age"])
User: derrick is 45
User: nancy is 94
User: joe is 35
# make sure to use parenthesis around FilterExpressions
# if initializing them while constructing them
age_range = (RedisNum("age") > 18) & (RedisNum("age") < 99)
results = rds.similarity_search("foo", filter=age_range)

for result in results:
print("User:", result.metadata["user"], "is", result.metadata["age"])
User: derrick is 45
User: nancy is 94
User: joe is 35

Redis as Retrieverโ€‹

Here we go over different options for using the vector store as a retriever.

There are three different search methods we can use to do retrieval. By default, it will use semantic similarity.

query = "foo"
results = rds.similarity_search_with_score(query, k=3, return_metadata=True)

for result in results:
print("Content:", result[0].page_content, " --- Score: ", result[1])
Content: foo  --- Score:  0.0
Content: foo --- Score: 0.0
Content: foo --- Score: 0.0
retriever = rds.as_retriever(search_type="similarity", search_kwargs={"k": 4})
docs = retriever.get_relevant_documents(query)
docs
[Document(page_content='foo', metadata={'id': 'doc:users_modified:988ecca7574048e396756efc0e79aeca', 'user': 'john', 'job': 'engineer', 'credit_score': 'high', 'age': '18'}),
Document(page_content='foo', metadata={'id': 'doc:users_modified:009b1afeb4084cc6bdef858c7a99b48e', 'user': 'derrick', 'job': 'doctor', 'credit_score': 'low', 'age': '45'}),
Document(page_content='foo', metadata={'id': 'doc:users_modified:7087cee9be5b4eca93c30fbdd09a2731', 'user': 'nancy', 'job': 'doctor', 'credit_score': 'high', 'age': '94'}),
Document(page_content='bar', metadata={'id': 'doc:users_modified:01ef6caac12b42c28ad870aefe574253', 'user': 'tyler', 'job': 'engineer', 'credit_score': 'high', 'age': '100'})]

There is also the similarity_distance_threshold retriever which allows the user to specify the vector distance

retriever = rds.as_retriever(
search_type="similarity_distance_threshold",
search_kwargs={"k": 4, "distance_threshold": 0.1},
)
docs = retriever.get_relevant_documents(query)
docs
[Document(page_content='foo', metadata={'id': 'doc:users_modified:988ecca7574048e396756efc0e79aeca', 'user': 'john', 'job': 'engineer', 'credit_score': 'high', 'age': '18'}),
Document(page_content='foo', metadata={'id': 'doc:users_modified:009b1afeb4084cc6bdef858c7a99b48e', 'user': 'derrick', 'job': 'doctor', 'credit_score': 'low', 'age': '45'}),
Document(page_content='foo', metadata={'id': 'doc:users_modified:7087cee9be5b4eca93c30fbdd09a2731', 'user': 'nancy', 'job': 'doctor', 'credit_score': 'high', 'age': '94'})]

Lastly, the similarity_score_threshold allows the user to define the minimum score for similar documents

retriever = rds.as_retriever(
search_type="similarity_score_threshold",
search_kwargs={"score_threshold": 0.9, "k": 10},
)
retriever.get_relevant_documents("foo")
[Document(page_content='foo', metadata={'id': 'doc:users_modified:988ecca7574048e396756efc0e79aeca', 'user': 'john', 'job': 'engineer', 'credit_score': 'high', 'age': '18'}),
Document(page_content='foo', metadata={'id': 'doc:users_modified:009b1afeb4084cc6bdef858c7a99b48e', 'user': 'derrick', 'job': 'doctor', 'credit_score': 'low', 'age': '45'}),
Document(page_content='foo', metadata={'id': 'doc:users_modified:7087cee9be5b4eca93c30fbdd09a2731', 'user': 'nancy', 'job': 'doctor', 'credit_score': 'high', 'age': '94'})]
retriever = rds.as_retriever(
search_type="mmr", search_kwargs={"fetch_k": 20, "k": 4, "lambda_mult": 0.1}
)
retriever.get_relevant_documents("foo")
[Document(page_content='foo', metadata={'id': 'doc:users:8f6b673b390647809d510112cde01a27', 'user': 'john', 'job': 'engineer', 'credit_score': 'high', 'age': '18'}),
Document(page_content='bar', metadata={'id': 'doc:users:93521560735d42328b48c9c6f6418d6a', 'user': 'tyler', 'job': 'engineer', 'credit_score': 'high', 'age': '100'}),
Document(page_content='foo', metadata={'id': 'doc:users:125ecd39d07845eabf1a699d44134a5b', 'user': 'nancy', 'job': 'doctor', 'credit_score': 'high', 'age': '94'}),
Document(page_content='foo', metadata={'id': 'doc:users:d6200ab3764c466082fde3eaab972a2a', 'user': 'derrick', 'job': 'doctor', 'credit_score': 'low', 'age': '45'})]

Delete keys and indexโ€‹

To delete your entries you have to address them by their keys.

Redis.delete(keys, redis_url="redis://localhost:6379")
True
# delete the indices too
Redis.drop_index(
index_name="users", delete_documents=True, redis_url="redis://localhost:6379"
)
Redis.drop_index(
index_name="users_modified",
delete_documents=True,
redis_url="redis://localhost:6379",
)
True