Redis use filters

[tarekabouzeid]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

retriever = redis.as_retriever(
            search_type="similarity_distance_threshold",
            search_kwargs="{'include_metadata': True,'distance_threshold': 0.8,'k': 5}"
        )

Description

• I am trying to pass some Redis filters here in redis retriever, but i always get malformated, can someone please provide an example of how to pass filters to redis retriever ?

System Info

langchain 0.0.346
langchain-core 0.0.10

python 3.9.18

[dosubot[bot]]

🤖

Hey there @tarekabouzeid! Great to see you diving into new challenges with LangChain. How's everything going on your end?

To pass filters to the Redis retriever in LangChain, you need to create instances of the RedisFilterField subclasses (RedisTag, RedisNum, RedisText) and use their methods to define the filter conditions. These instances can then be combined using logical operators to create complex filter expressions.

Here are some examples:

1. Filtering on a tag field:

from langchain_community.vectorstores.redis import RedisTag

# Create a filter where the 'brand' tag is 'nike'
filter = RedisTag("brand") == "nike"

2. Filtering on a numeric field:

from langchain_community.vectorstores.redis import RedisNum

# Create a filter where the 'age' is greater than 18
filter = RedisNum("age") > 18

3. Filtering on a text field:

from langchain_community.vectorstores.redis import RedisText

# Create a filter where the 'job' field contains the term 'engineer'
filter = RedisText("job") % "engineer"

4. Combining filters:

from langchain_community.vectorstores.redis import RedisTag, RedisNum

# Create a filter where the 'brand' tag is 'nike' and the 'price' is less than 100
filter = (RedisTag("brand") == "nike") & (RedisNum("price") < 100)

The resulting filter objects can be passed to the Redis retriever. The exact method of passing the filter will depend on the specific retriever's API. You can refer to the filters.py file in the LangChain repository for more details.

Regarding your code snippet, it seems like you're trying to pass a string as search_kwargs. Instead, you should pass a dictionary. Here's how you can do it:

retriever = redis.as_retriever(
    search_type="similarity_distance_threshold",
    search_kwargs={'include_metadata': True, 'distance_threshold': 0.8, 'k': 5}
)

I hope this helps! If you have any other questions, feel free to ask.

Sources

• libs/community/langchain_community/vectorstores/redis/filters.py
• libs/langchain/langchain/vectorstores/redis/base.py

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[tarekabouzeid]

how to pass this redis filter to redis.as_retriever then ?