mirror of https://github.com/hwchase17/langchain
Harrison/myscale self query (#6376)
Co-authored-by: Fangrui Liu <fangruil@moqi.ai> Co-authored-by: 刘 方瑞 <fangrui.liu@outlook.com> Co-authored-by: Fangrui.Liu <fangrui.liu@ubc.ca>pull/6390/head
parent
bd8d418a95
commit
9bf5b0defa
@ -0,0 +1,370 @@
|
||||
{
|
||||
"cells": [
|
||||
{
|
||||
"attachments": {},
|
||||
"cell_type": "markdown",
|
||||
"id": "13afcae7",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"# Self-querying with MyScale\n",
|
||||
"\n",
|
||||
">[MyScale](https://docs.myscale.com/en/) is an integrated vector database. You can access your database in SQL and also from here, LangChain. MyScale can make a use of [various data types and functions for filters](https://blog.myscale.com/2023/06/06/why-integrated-database-solution-can-boost-your-llm-apps/#filter-on-anything-without-constraints). It will boost up your LLM app no matter if you are scaling up your data or expand your system to broader application.\n",
|
||||
"\n",
|
||||
"In the notebook we'll demo the `SelfQueryRetriever` wrapped around a MyScale vector store with some extra piece we contributed to LangChain. In short, it can be concluded into 4 points:\n",
|
||||
"1. Add `contain` comparator to match list of any if there is more than one element matched\n",
|
||||
"2. Add `timestamp` data type for datetime match (ISO-format, or YYYY-MM-DD)\n",
|
||||
"3. Add `like` comparator for string pattern search\n",
|
||||
"4. Add arbitrary function capability"
|
||||
]
|
||||
},
|
||||
{
|
||||
"attachments": {},
|
||||
"cell_type": "markdown",
|
||||
"id": "68e75fb9",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Creating a MyScale vectorstore\n",
|
||||
"MyScale has already been integrated to LangChain for a while. So you can follow [this notebook](../../vectorstores/examples/myscale.ipynb) to create your own vectorstore for a self-query retriever.\n",
|
||||
"\n",
|
||||
"NOTE: All self-query retrievers requires you to have `lark` installed (`pip install lark`). We use `lark` for grammar definition. Before you proceed to the next step, we also want to remind you that `clickhouse-connect` is also needed to interact with your MyScale backend."
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "63a8af5b",
|
||||
"metadata": {
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"! pip install lark clickhouse-connect"
|
||||
]
|
||||
},
|
||||
{
|
||||
"attachments": {},
|
||||
"cell_type": "markdown",
|
||||
"id": "83811610-7df3-4ede-b268-68a6a83ba9e2",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"In this tutorial we follow other example's setting and use `OpenAIEmbeddings`. Remember to get a OpenAI API Key for valid accesss to LLMs."
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "dd01b61b-7d32-4a55-85d6-b2d2d4f18840",
|
||||
"metadata": {
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"import os\n",
|
||||
"import getpass\n",
|
||||
"\n",
|
||||
"os.environ['OPENAI_API_KEY'] = getpass.getpass('OpenAI API Key:')\n",
|
||||
"os.environ['MYSCALE_HOST'] = getpass.getpass('MyScale URL:')\n",
|
||||
"os.environ['MYSCALE_PORT'] = getpass.getpass('MyScale Port:')\n",
|
||||
"os.environ['MYSCALE_USERNAME'] = getpass.getpass('MyScale Username:')\n",
|
||||
"os.environ['MYSCALE_PASSWORD'] = getpass.getpass('MyScale Password:')"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "cb4a5787",
|
||||
"metadata": {
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"from langchain.schema import Document\n",
|
||||
"from langchain.embeddings.openai import OpenAIEmbeddings\n",
|
||||
"from langchain.vectorstores import MyScale\n",
|
||||
"\n",
|
||||
"embeddings = OpenAIEmbeddings()"
|
||||
]
|
||||
},
|
||||
{
|
||||
"attachments": {},
|
||||
"cell_type": "markdown",
|
||||
"id": "bf7f6fc4",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Create some sample data\n",
|
||||
"As you can see, the data we created has some difference to other self-query retrievers. We replaced keyword `year` to `date` which gives you a finer control on timestamps. We also altered the type of keyword `gerne` to list of strings, where LLM can use a new `contain` comparator to construct filters. We also provides comparator `like` and arbitrary function support to filters, which will be introduced in next few cells.\n",
|
||||
"\n",
|
||||
"Now let's look at the data first."
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "bcbe04d9",
|
||||
"metadata": {
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"docs = [\n",
|
||||
" Document(page_content=\"A bunch of scientists bring back dinosaurs and mayhem breaks loose\", metadata={\"date\": \"1993-07-02\", \"rating\": 7.7, \"genre\": [\"science fiction\"]}),\n",
|
||||
" Document(page_content=\"Leo DiCaprio gets lost in a dream within a dream within a dream within a ...\", metadata={\"date\": \"2010-12-30\", \"director\": \"Christopher Nolan\", \"rating\": 8.2}),\n",
|
||||
" Document(page_content=\"A psychologist / detective gets lost in a series of dreams within dreams within dreams and Inception reused the idea\", metadata={\"date\": \"2006-04-23\", \"director\": \"Satoshi Kon\", \"rating\": 8.6}),\n",
|
||||
" Document(page_content=\"A bunch of normal-sized women are supremely wholesome and some men pine after them\", metadata={\"date\": \"2019-08-22\", \"director\": \"Greta Gerwig\", \"rating\": 8.3}),\n",
|
||||
" Document(page_content=\"Toys come alive and have a blast doing so\", metadata={\"date\": \"1995-02-11\", \"genre\": [\"animated\"]}),\n",
|
||||
" Document(page_content=\"Three men walk into the Zone, three men walk out of the Zone\", metadata={\"date\": \"1979-09-10\", \"rating\": 9.9, \"director\": \"Andrei Tarkovsky\", \"genre\": [\"science fiction\", \"adventure\"], \"rating\": 9.9})\n",
|
||||
"]\n",
|
||||
"vectorstore = MyScale.from_documents(\n",
|
||||
" docs, \n",
|
||||
" embeddings, \n",
|
||||
")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"attachments": {},
|
||||
"cell_type": "markdown",
|
||||
"id": "5ecaab6d",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Creating our self-querying retriever\n",
|
||||
"Just like other retrievers... Simple and nice."
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "86e34dbf",
|
||||
"metadata": {
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"from langchain.llms import OpenAI\n",
|
||||
"from langchain.retrievers.self_query.base import SelfQueryRetriever\n",
|
||||
"from langchain.chains.query_constructor.base import AttributeInfo\n",
|
||||
"\n",
|
||||
"metadata_field_info=[\n",
|
||||
" AttributeInfo(\n",
|
||||
" name=\"genre\",\n",
|
||||
" description=\"The genres of the movie\", \n",
|
||||
" type=\"list[string]\", \n",
|
||||
" ),\n",
|
||||
" # If you want to include length of a list, just define it as a new column\n",
|
||||
" # This will teach the LLM to use it as a column when constructing filter.\n",
|
||||
" AttributeInfo(\n",
|
||||
" name=\"length(genre)\",\n",
|
||||
" description=\"The lenth of genres of the movie\", \n",
|
||||
" type=\"integer\", \n",
|
||||
" ),\n",
|
||||
" # Now you can define a column as timestamp. By simply set the type to timestamp.\n",
|
||||
" AttributeInfo(\n",
|
||||
" name=\"date\",\n",
|
||||
" description=\"The date the movie was released\", \n",
|
||||
" type=\"timestamp\", \n",
|
||||
" ),\n",
|
||||
" AttributeInfo(\n",
|
||||
" name=\"director\",\n",
|
||||
" description=\"The name of the movie director\", \n",
|
||||
" type=\"string\", \n",
|
||||
" ),\n",
|
||||
" AttributeInfo(\n",
|
||||
" name=\"rating\",\n",
|
||||
" description=\"A 1-10 rating for the movie\",\n",
|
||||
" type=\"float\"\n",
|
||||
" ),\n",
|
||||
"]\n",
|
||||
"document_content_description = \"Brief summary of a movie\"\n",
|
||||
"llm = OpenAI(temperature=0)\n",
|
||||
"retriever = SelfQueryRetriever.from_llm(llm, vectorstore, document_content_description, metadata_field_info, verbose=True)"
|
||||
]
|
||||
},
|
||||
{
|
||||
"attachments": {},
|
||||
"cell_type": "markdown",
|
||||
"id": "ea9df8d4",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Testing it out with self-query retriever's existing functionalities\n",
|
||||
"And now we can try actually using our retriever!"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "38a126e9",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# This example only specifies a relevant query\n",
|
||||
"retriever.get_relevant_documents(\"What are some movies about dinosaurs\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "fc3f1e6e",
|
||||
"metadata": {
|
||||
"scrolled": false
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# This example only specifies a filter\n",
|
||||
"retriever.get_relevant_documents(\"I want to watch a movie rated higher than 8.5\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "b19d4da0",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# This example specifies a query and a filter\n",
|
||||
"retriever.get_relevant_documents(\"Has Greta Gerwig directed any movies about women\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "f900e40e",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# This example specifies a composite filter\n",
|
||||
"retriever.get_relevant_documents(\"What's a highly rated (above 8.5) science fiction film?\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "12a51522",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# This example specifies a query and composite filter\n",
|
||||
"retriever.get_relevant_documents(\"What's a movie after 1990 but before 2005 that's all about toys, and preferably is animated\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"attachments": {},
|
||||
"cell_type": "markdown",
|
||||
"id": "86371ac8",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"# Wait a second... What else?\n",
|
||||
"\n",
|
||||
"Self-query retriever with MyScale can do more! Let's find out."
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "1d043096",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# You can use length(genres) to do anything you want\n",
|
||||
"retriever.get_relevant_documents(\"What's a movie that have more than 1 genres?\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "d570d33c",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# Fine-grained datetime? You got it already.\n",
|
||||
"retriever.get_relevant_documents(\"What's a movie that release after feb 1995?\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "fbe0b21b",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# Don't know what your exact filter should be? Use string pattern match!\n",
|
||||
"retriever.get_relevant_documents(\"What's a movie whose name is like Andrei?\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "6a514104",
|
||||
"metadata": {},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# Contain works for lists: so you can match a list with contain comparator!\n",
|
||||
"retriever.get_relevant_documents(\"What's a movie who has genres science fiction and adventure?\")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"attachments": {},
|
||||
"cell_type": "markdown",
|
||||
"id": "39bd1de1-b9fe-4a98-89da-58d8a7a6ae51",
|
||||
"metadata": {},
|
||||
"source": [
|
||||
"## Filter k\n",
|
||||
"\n",
|
||||
"We can also use the self query retriever to specify `k`: the number of documents to fetch.\n",
|
||||
"\n",
|
||||
"We can do this by passing `enable_limit=True` to the constructor."
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "bff36b88-b506-4877-9c63-e5a1a8d78e64",
|
||||
"metadata": {
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"retriever = SelfQueryRetriever.from_llm(\n",
|
||||
" llm, \n",
|
||||
" vectorstore, \n",
|
||||
" document_content_description, \n",
|
||||
" metadata_field_info, \n",
|
||||
" enable_limit=True,\n",
|
||||
" verbose=True\n",
|
||||
")"
|
||||
]
|
||||
},
|
||||
{
|
||||
"cell_type": "code",
|
||||
"execution_count": null,
|
||||
"id": "2758d229-4f97-499c-819f-888acaf8ee10",
|
||||
"metadata": {
|
||||
"tags": []
|
||||
},
|
||||
"outputs": [],
|
||||
"source": [
|
||||
"# This example only specifies a relevant query\n",
|
||||
"retriever.get_relevant_documents(\"what are two movies about dinosaurs\")"
|
||||
]
|
||||
}
|
||||
],
|
||||
"metadata": {
|
||||
"kernelspec": {
|
||||
"display_name": "Python 3 (ipykernel)",
|
||||
"language": "python",
|
||||
"name": "python3"
|
||||
},
|
||||
"language_info": {
|
||||
"codemirror_mode": {
|
||||
"name": "ipython",
|
||||
"version": 3
|
||||
},
|
||||
"file_extension": ".py",
|
||||
"mimetype": "text/x-python",
|
||||
"name": "python",
|
||||
"nbconvert_exporter": "python",
|
||||
"pygments_lexer": "ipython3",
|
||||
"version": "3.8.8"
|
||||
}
|
||||
},
|
||||
"nbformat": 4,
|
||||
"nbformat_minor": 5
|
||||
}
|
@ -0,0 +1,106 @@
|
||||
import datetime
|
||||
import re
|
||||
from typing import Any, Callable, Dict, Tuple
|
||||
|
||||
from langchain.chains.query_constructor.ir import (
|
||||
Comparator,
|
||||
Comparison,
|
||||
Operation,
|
||||
Operator,
|
||||
StructuredQuery,
|
||||
Visitor,
|
||||
)
|
||||
|
||||
|
||||
def DEFAULT_COMPOSER(op_name: str) -> Callable:
|
||||
def f(*args: Any) -> str:
|
||||
args_: map[str] = map(str, args)
|
||||
return f" {op_name} ".join(args_)
|
||||
|
||||
return f
|
||||
|
||||
|
||||
def FUNCTION_COMPOSER(op_name: str) -> Callable:
|
||||
def f(*args: Any) -> str:
|
||||
args_: map[str] = map(str, args)
|
||||
return f"{op_name}({','.join(args_)})"
|
||||
|
||||
return f
|
||||
|
||||
|
||||
class MyScaleTranslator(Visitor):
|
||||
"""Logic for converting internal query language elements to valid filters."""
|
||||
|
||||
allowed_operators = [Operator.AND, Operator.OR, Operator.NOT]
|
||||
"""Subset of allowed logical operators."""
|
||||
|
||||
allowed_comparators = [
|
||||
Comparator.EQ,
|
||||
Comparator.GT,
|
||||
Comparator.GTE,
|
||||
Comparator.LT,
|
||||
Comparator.LTE,
|
||||
Comparator.CONTAIN,
|
||||
Comparator.LIKE,
|
||||
]
|
||||
|
||||
map_dict = {
|
||||
Operator.AND: DEFAULT_COMPOSER("AND"),
|
||||
Operator.OR: DEFAULT_COMPOSER("OR"),
|
||||
Operator.NOT: DEFAULT_COMPOSER("NOT"),
|
||||
Comparator.EQ: DEFAULT_COMPOSER("="),
|
||||
Comparator.GT: DEFAULT_COMPOSER(">"),
|
||||
Comparator.GTE: DEFAULT_COMPOSER(">="),
|
||||
Comparator.LT: DEFAULT_COMPOSER("<"),
|
||||
Comparator.LTE: DEFAULT_COMPOSER("<="),
|
||||
Comparator.CONTAIN: FUNCTION_COMPOSER("has"),
|
||||
Comparator.LIKE: DEFAULT_COMPOSER("ILIKE"),
|
||||
}
|
||||
|
||||
def __init__(self, metadata_key: str = "metadata") -> None:
|
||||
super().__init__()
|
||||
self.metadata_key = metadata_key
|
||||
|
||||
def visit_operation(self, operation: Operation) -> Dict:
|
||||
args = [arg.accept(self) for arg in operation.arguments]
|
||||
func = operation.operator
|
||||
self._validate_func(func)
|
||||
return self.map_dict[func](*args)
|
||||
|
||||
def visit_comparison(self, comparison: Comparison) -> Dict:
|
||||
regex = "\((.*?)\)"
|
||||
matched = re.search("\(\w+\)", comparison.attribute)
|
||||
|
||||
# If arbitrary function is applied to an attribute
|
||||
if matched:
|
||||
attr = re.sub(
|
||||
regex,
|
||||
f"({self.metadata_key}.{matched.group(0)[1:-1]})",
|
||||
comparison.attribute,
|
||||
)
|
||||
else:
|
||||
attr = f"{self.metadata_key}.{comparison.attribute}"
|
||||
value = comparison.value
|
||||
comp = comparison.comparator
|
||||
|
||||
value = f"'{value}'" if type(value) is str else value
|
||||
|
||||
# convert timestamp for datetime objects
|
||||
if type(value) is datetime.date:
|
||||
attr = f"parseDateTime32BestEffort({attr})"
|
||||
value = f"parseDateTime32BestEffort('{value.strftime('%Y-%m-%d')}')"
|
||||
|
||||
# string pattern match
|
||||
if comp is Comparator.LIKE:
|
||||
value = f"'%{value[1:-1]}%'"
|
||||
return self.map_dict[comp](attr, value)
|
||||
|
||||
def visit_structured_query(
|
||||
self, structured_query: StructuredQuery
|
||||
) -> Tuple[str, dict]:
|
||||
print(structured_query)
|
||||
if structured_query.filter is None:
|
||||
kwargs = {}
|
||||
else:
|
||||
kwargs = {"where_str": structured_query.filter.accept(self)}
|
||||
return structured_query.query, kwargs
|
@ -0,0 +1,44 @@
|
||||
from typing import Any, Tuple
|
||||
|
||||
import pytest
|
||||
|
||||
from langchain.chains.query_constructor.ir import (
|
||||
Comparator,
|
||||
Comparison,
|
||||
Operation,
|
||||
Operator,
|
||||
)
|
||||
from langchain.retrievers.self_query.myscale import MyScaleTranslator
|
||||
|
||||
DEFAULT_TRANSLATOR = MyScaleTranslator()
|
||||
|
||||
|
||||
@pytest.mark.parametrize(
|
||||
"triplet",
|
||||
[
|
||||
(Comparator.LT, 2, "metadata.foo < 2"),
|
||||
(Comparator.LTE, 2, "metadata.foo <= 2"),
|
||||
(Comparator.GT, 2, "metadata.foo > 2"),
|
||||
(Comparator.GTE, 2, "metadata.foo >= 2"),
|
||||
(Comparator.CONTAIN, 2, "has(metadata.foo,2)"),
|
||||
(Comparator.LIKE, "bar", "metadata.foo ILIKE '%bar%'"),
|
||||
],
|
||||
)
|
||||
def test_visit_comparison(triplet: Tuple[Comparator, Any, str]) -> None:
|
||||
comparator, value, expected = triplet
|
||||
comp = Comparison(comparator=comparator, attribute="foo", value=value)
|
||||
actual = DEFAULT_TRANSLATOR.visit_comparison(comp)
|
||||
assert expected == actual
|
||||
|
||||
|
||||
def test_visit_operation() -> None:
|
||||
op = Operation(
|
||||
operator=Operator.AND,
|
||||
arguments=[
|
||||
Comparison(comparator=Comparator.LT, attribute="foo", value=2),
|
||||
Comparison(comparator=Comparator.EQ, attribute="bar", value="baz"),
|
||||
],
|
||||
)
|
||||
expected = "metadata.foo < 2 AND metadata.bar = 'baz'"
|
||||
actual = DEFAULT_TRANSLATOR.visit_operation(op)
|
||||
assert expected == actual
|
Loading…
Reference in New Issue