知识图谱 RAG 查询引擎¶
Graph RAG¶
Graph RAG 是一种知识增强型 RAG 方法，用于从知识图谱中检索与给定任务相关的信息。通常，这是基于与任务相关的实体的子图构建上下文。
GraphStore 支持的 RAG vs VectorStore RAG¶
正如我们在本教程中比较了 Graph RAG 在某些用例中的帮助作用，结果显示，知识图谱作为独特的信息格式，可以缓解由“分割和嵌入”RAG 方法的本质引起的一些问题。
为什么选择知识图谱 RAG 查询引擎¶
在 Llama Index 中，有两种情况可以应用 Graph RAG
- 使用 Llama Index，通过 LLM 甚至本地模型从文档构建知识图谱，要做到这一点，我们应该选择
KnowledgeGraphIndex。 - 利用现有的知识图谱，在这种情况下，我们应该使用
KnowledgeGraphRAGQueryEngine。
注意，Llama Index 中与 KG 相关的第三种查询引擎是
NL2GraphQuery或Text2Cypher，无论是否存在 KG，都可以通过KnowledgeGraphQueryEngine来完成。
在我们开始知识图谱 RAG 查询引擎演示之前，让我们先准备好 Llama Index 的基本环境。
如果您在 Colab 上打开这个 Notebook，您可能需要安装 LlamaIndex 🦙。
In [ ]
已复制！
%pip install llama-index-llms-azure-openai
%pip install llama-index-graph-stores-nebula
%pip install llama-index-llms-openai
%pip install llama-index-embeddings-azure-openai
%pip install llama-index-llms-azure-openai %pip install llama-index-graph-stores-nebula %pip install llama-index-llms-openai %pip install llama-index-embeddings-azure-openai
In [ ]
已复制！
!pip install llama-index
!pip install llama-index
OpenAI¶
In [ ]
已复制！
# For OpenAI
import os
os.environ["OPENAI_API_KEY"] = "sk-..."
import logging
import sys
logging.basicConfig(
stream=sys.stdout, level=logging.INFO
) # logging.DEBUG for more verbose output
# define LLM
from llama_index.llms.openai import OpenAI
from llama_index.core import Settings
Settings.llm = OpenAI(temperature=0, model="gpt-3.5-turbo")
Settings.chunk_size = 512
# 对于 OpenAI import os os.environ["OPENAI_API_KEY"] = "sk-..." import logging import sys logging.basicConfig( stream=sys.stdout, level=logging.INFO ) # 使用 logging.DEBUG 获取更详细的输出 # 定义 LLM from llama_index.llms.openai import OpenAI from llama_index.core import Settings Settings.llm = OpenAI(temperature=0, model="gpt-3.5-turbo") Settings.chunk_size = 512
Azure¶
In [ ]
已复制！
from llama_index.llms.azure_openai import AzureOpenAI
from llama_index.embeddings.azure_openai import AzureOpenAIEmbedding
# For Azure OpenAI
api_key = "<api-key>"
azure_endpoint = "https://<your-resource-name>.openai.azure.com/"
api_version = "2023-07-01-preview"
llm = AzureOpenAI(
model="gpt-35-turbo-16k",
deployment_name="my-custom-llm",
api_key=api_key,
azure_endpoint=azure_endpoint,
api_version=api_version,
)
# You need to deploy your own embedding model as well as your own chat completion model
embed_model = AzureOpenAIEmbedding(
model="text-embedding-ada-002",
deployment_name="my-custom-embedding",
api_key=api_key,
azure_endpoint=azure_endpoint,
api_version=api_version,
)
from llama_index.llms.azure_openai import AzureOpenAI from llama_index.embeddings.azure_openai import AzureOpenAIEmbedding # 对于 Azure OpenAI api_key = "" azure_endpoint = "https://.openai.azure.com/" api_version = "2023-07-01-preview" llm = AzureOpenAI( model="gpt-35-turbo-16k", deployment_name="my-custom-llm", api_key=api_key, azure_endpoint=azure_endpoint, api_version=api_version, ) # 您需要部署自己的嵌入模型和聊天完成模型 embed_model = AzureOpenAIEmbedding( model="text-embedding-ada-002", deployment_name="my-custom-embedding", api_key=api_key, azure_endpoint=azure_endpoint, api_version=api_version, )
In [ ]
已复制！
from llama_index.core import Settings
Settings.llm = llm
Settings.embed_model = embed_model
Settings.chunk_size = 512
from llama_index.core import Settings Settings.llm = llm Settings.embed_model = embed_model Settings.chunk_size = 512
为 NebulaGraph 做准备¶
在此演示中，我们以 NebulaGraphStore 为例，因此在下一步对现有 KG 执行 Graph RAG 之前，请确保您有一个正在运行且已定义数据 schema 的 NebulaGraph。
此步骤安装 NebulaGraph 客户端，并准备定义 NebulaGraph 图空间的上下文。
In [ ]
已复制！
# Create a NebulaGraph (version 3.5.0 or newer) cluster with:
# Option 0 for machines with Docker: `curl -fsSL nebula-up.siwei.io/install.sh | bash`
# Option 1 for Desktop: NebulaGraph Docker Extension https://hub.docker.com/extensions/weygu/nebulagraph-dd-ext
# If not, create it with the following commands from NebulaGraph's console:
# CREATE SPACE llamaindex(vid_type=FIXED_STRING(256), partition_num=1, replica_factor=1);
# :sleep 10;
# USE llamaindex;
# CREATE TAG entity(name string);
# CREATE EDGE relationship(relationship string);
# :sleep 10;
# CREATE TAG INDEX entity_index ON entity(name(256));
%pip install ipython-ngql nebula3-python
os.environ["NEBULA_USER"] = "root"
os.environ["NEBULA_PASSWORD"] = "nebula" # default is "nebula"
os.environ[
"NEBULA_ADDRESS"
] = "127.0.0.1:9669" # assumed we have NebulaGraph installed locally
space_name = "llamaindex"
edge_types, rel_prop_names = ["relationship"], [
"relationship"
] # default, could be omit if create from an empty kg
tags = ["entity"] # default, could be omit if create from an empty kg
# 使用以下方式创建 NebulaGraph (版本 3.5.0 或更高) 集群： # 方案 0 适用于带有 Docker 的机器：`curl -fsSL nebula-up.siwei.io/install.sh | bash` # 方案 1 适用于桌面：NebulaGraph Docker 扩展 https://hub.docker.com/extensions/weygu/nebulagraph-dd-ext # 如果尚未创建，请使用 NebulaGraph 控制台中的以下命令创建： # CREATE SPACE llamaindex(vid_type=FIXED_STRING(256), partition_num=1, replica_factor=1); # :sleep 10; # USE llamaindex; # CREATE TAG entity(name string); # CREATE EDGE relationship(relationship string); # :sleep 10; # CREATE TAG INDEX entity_index ON entity(name(256)); %pip install ipython-ngql nebula3-python os.environ["NEBULA_USER"] = "root" os.environ["NEBULA_PASSWORD"] = "nebula" # 默认为 "nebula" os.environ[ "NEBULA_ADDRESS" ] = "127.0.0.1:9669" # 假设我们在本地安装了 NebulaGraph space_name = "llamaindex" edge_types, rel_prop_names = ["relationship"], [ "relationship" ] # 默认值，如果从一个空的知识图谱创建，则可以省略 tags = ["entity"] # 默认值，如果从一个空的知识图谱创建，则可以省略
Requirement already satisfied: ipython-ngql in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (0.5)
Requirement already satisfied: nebula3-python in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (3.4.0)
Requirement already satisfied: Jinja2 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from ipython-ngql) (3.1.2)
Requirement already satisfied: pandas in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from ipython-ngql) (2.0.3)
Requirement already satisfied: httplib2>=0.20.0 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from nebula3-python) (0.22.0)
Requirement already satisfied: six>=1.16.0 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from nebula3-python) (1.16.0)
Requirement already satisfied: pytz>=2021.1 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from nebula3-python) (2023.3)
Requirement already satisfied: future>=0.18.0 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from nebula3-python) (0.18.3)
Requirement already satisfied: pyparsing!=3.0.0,!=3.0.1,!=3.0.2,!=3.0.3,<4,>=2.4.2 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from httplib2>=0.20.0->nebula3-python) (3.0.9)
Requirement already satisfied: MarkupSafe>=2.0 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from Jinja2->ipython-ngql) (2.1.3)
Requirement already satisfied: numpy>=1.20.3 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from pandas->ipython-ngql) (1.25.2)
Requirement already satisfied: tzdata>=2022.1 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from pandas->ipython-ngql) (2023.3)
Requirement already satisfied: python-dateutil>=2.8.2 in /Users/loganmarkewich/llama_index/llama-index/lib/python3.9/site-packages (from pandas->ipython-ngql) (2.8.2)
WARNING: You are using pip version 21.2.4; however, version 23.2.1 is available.
You should consider upgrading via the '/Users/loganmarkewich/llama_index/llama-index/bin/python -m pip install --upgrade pip' command.
Note: you may need to restart the kernel to use updated packages.
然后我们可以实例化一个 NebulaGraphStore，以便将其创建为 StorageContext 的 graph_store。
In [ ]
已复制！
from llama_index.core import StorageContext
from llama_index.graph_stores.nebula import NebulaGraphStore
graph_store = NebulaGraphStore(
space_name=space_name,
edge_types=edge_types,
rel_prop_names=rel_prop_names,
tags=tags,
)
storage_context = StorageContext.from_defaults(graph_store=graph_store)
from llama_index.core import StorageContext from llama_index.graph_stores.nebula import NebulaGraphStore graph_store = NebulaGraphStore( space_name=space_name, edge_types=edge_types, rel_prop_names=rel_prop_names, tags=tags, ) storage_context = StorageContext.from_defaults(graph_store=graph_store)
这里，我们假设拥有与本教程中相同的知识图谱
执行 Graph RAG 查询¶
最后，让我们演示如何对现有知识图谱执行 Graph RAG。
我们只需要使用 RetrieverQueryEngine 并将其检索器配置为 KnowledgeGraphRAGRetriever。
KnowledgeGraphRAGRetriever 执行以下步骤
- 搜索问题/任务的相关实体
- 从 KG 获取这些实体的子图（默认深度为 2）
- 基于子图构建上下文
请注意，搜索相关实体的方式可以是基于关键词提取或基于嵌入，这由 KnowledgeGraphRAGRetriever 的参数 retriever_mode 控制，支持的选项有
- "keyword"（关键词）
- "embedding"（嵌入，尚未实现）
- "keyword_embedding"（关键词嵌入，尚未实现）
以下是关于如何使用 RetrieverQueryEngine 和 KnowledgeGraphRAGRetriever 的示例
In [ ]
已复制！
from llama_index.core.query_engine import RetrieverQueryEngine
from llama_index.core.retrievers import KnowledgeGraphRAGRetriever
graph_rag_retriever = KnowledgeGraphRAGRetriever(
storage_context=storage_context,
verbose=True,
)
query_engine = RetrieverQueryEngine.from_args(
graph_rag_retriever,
)
from llama_index.core.query_engine import RetrieverQueryEngine from llama_index.core.retrievers import KnowledgeGraphRAGRetriever graph_rag_retriever = KnowledgeGraphRAGRetriever( storage_context=storage_context, verbose=True, ) query_engine = RetrieverQueryEngine.from_args( graph_rag_retriever, )
然后我们可以像这样查询它
In [ ]
已复制！
from IPython.display import display, Markdown
response = query_engine.query(
"Tell me about Peter Quill?",
)
display(Markdown(f"<b>{response}</b>"))
from IPython.display import display, Markdown response = query_engine.query( "Tell me about Peter Quill?", ) display(Markdown(f"{response}"))
Entities processed: ['Star', 'Lord', 'Marvel', 'Quill', 'Galaxy', 'Guardians', 'Guardians of the Galaxy', 'Star-Lord', 'Peter Quill', 'Peter'] Entities processed: ['Star', 'Lord', 'Marvel', 'Quill', 'Galaxy', 'Guardians', 'Guardians of the Galaxy', 'Star-Lord', 'Peter Quill', 'Peter'] Graph RAG context: The following are knowledge sequence in max depth 2 in the form of `subject predicate, object, predicate_next_hop, object_next_hop ...` extracted based on key entities as subject: Guardians, is member of, Guardians, was experimented on, by the High Evolutionary Guardians, is member of, Guardians, considered to tell, origins Guardians, is member of, Guardians, origins, team-up movie Guardians, is member of, Guardians, befriended, his fellow Batch 89 test subjects Guardians, is member of, Guardians, sought to enhance and anthropomorphize animal lifeforms, to create an ideal society Guardians, is member of, Guardians, is creator of, Rocket Guardians, is member of, Guardians, is, Mantis Guardians, is member of, Guardians, is half-sister of, Mantis Guardians, is member of, Guardians, is, Kraglin Guardians, is member of, Guardians, developed psionic abilities, after being abandoned in outer space Guardians, is member of, Guardians, would portray, Cosmo Guardians, is member of, Guardians, recalls, his past Guardians, is member of, Guardians Guardians, is member of, Guardians, focus on, third Guardians-centric film Guardians, is member of, Guardians, is, Rocket Guardians, is member of, Guardians, backstory, flashbacks Guardians, is member of, Guardians, is former second-in-command of, Ravagers Quill, is half-sister of, Mantis, is member of, Guardians Quill, is half-sister of, Mantis, is, Mantis Quill, is in a state of depression, following the appearance of a variant of his dead lover Gamora Quill, is half-sister of, Mantis Peter Quill, is leader of, Guardians of the Galaxy, is sequel to, Guardians of the Galaxy Peter Quill, was raised by, a group of alien thieves and smugglers Peter Quill, would return to the MCU, May 2021 Peter Quill, is leader of, Guardians of the Galaxy Peter Quill, is half-human, half-Celestial Peter Quill, was abducted from Earth, as a child Guardians of the Galaxy, is sequel to, Guardians of the Galaxy, released in, Dolby Cinema Guardians of the Galaxy, is sequel to, Guardians of the Galaxy, released on, Disney+ Guardians of the Galaxy, is sequel to, Guardians of the Galaxy, is sequel to, Guardians of the Galaxy Vol. 2
彼得·奎尔（Peter Quill）是银河护卫队的领袖，也是《银河护卫队》系列电影的主角。他从小被一群外星窃贼和走私犯抚养长大，并在童年时期从地球被绑架。他是半人半天神，拥有挥舞一种名为“无限宝石”的能量武器的能力。他定于 2021 年 5 月回归 MCU。
In [ ]
已复制！
response = await query_engine.aquery(
"Tell me about Peter Quill?",
)
display(Markdown(f"<b>{response}</b>"))
response = await query_engine.aquery( "Tell me about Peter Quill?", ) display(Markdown(f"{response}"))
INFO:openai:message='OpenAI API response' path=https://api.openai.com/v1/completions processing_ms=611 request_id=1c07a89e18f19ac7bbc508507c2902d9 response_code=200 Entities processed: ['Star', 'Lord', 'Marvel', 'Quill', 'Galaxy', 'Guardians', 'Guardians of the Galaxy', 'Star-Lord', 'Peter Quill', 'Peter'] INFO:openai:message='OpenAI API response' path=https://api.openai.com/v1/completions processing_ms=992 request_id=6517cb63da3364acd33e816a9b3ee242 response_code=200 Entities processed: ['Star', 'Lord', 'Marvel', 'Quill', 'Galaxy', 'Guardians', 'Guardians of the Galaxy', 'Star-Lord', 'Peter Quill', 'Peter'] Graph RAG context: The following are knowledge sequence in max depth 2 in the form of `subject predicate, object, predicate_next_hop, object_next_hop ...` extracted based on key entities as subject: Guardians, is member of, Guardians, was experimented on, by the High Evolutionary Guardians, is member of, Guardians, considered to tell, origins Guardians, is member of, Guardians, origins, team-up movie Guardians, is member of, Guardians, befriended, his fellow Batch 89 test subjects Guardians, is member of, Guardians, sought to enhance and anthropomorphize animal lifeforms, to create an ideal society Guardians, is member of, Guardians, is creator of, Rocket Guardians, is member of, Guardians, is, Mantis Guardians, is member of, Guardians, is half-sister of, Mantis Guardians, is member of, Guardians, is, Kraglin Guardians, is member of, Guardians, developed psionic abilities, after being abandoned in outer space Guardians, is member of, Guardians, would portray, Cosmo Guardians, is member of, Guardians, recalls, his past Guardians, is member of, Guardians Guardians, is member of, Guardians, focus on, third Guardians-centric film Guardians, is member of, Guardians, is, Rocket Guardians, is member of, Guardians, backstory, flashbacks Guardians, is member of, Guardians, is former second-in-command of, Ravagers Quill, is half-sister of, Mantis, is member of, Guardians Quill, is half-sister of, Mantis, is, Mantis Quill, is in a state of depression, following the appearance of a variant of his dead lover Gamora Quill, is half-sister of, Mantis Peter Quill, is leader of, Guardians of the Galaxy, is sequel to, Guardians of the Galaxy Peter Quill, was raised by, a group of alien thieves and smugglers Peter Quill, would return to the MCU, May 2021 Peter Quill, is leader of, Guardians of the Galaxy Peter Quill, is half-human, half-Celestial Peter Quill, was abducted from Earth, as a child Guardians of the Galaxy, is sequel to, Guardians of the Galaxy, released in, Dolby Cinema Guardians of the Galaxy, is sequel to, Guardians of the Galaxy, released on, Disney+ Guardians of the Galaxy, is sequel to, Guardians of the Galaxy, is sequel to, Guardians of the Galaxy Vol. 2 INFO:openai:message='OpenAI API response' path=https://api.openai.com/v1/completions processing_ms=2384 request_id=b5a7e601affa751fbc7f957f3359a238 response_code=200
彼得·奎尔（Peter Quill）是银河护卫队的领袖，也是《银河护卫队》系列电影的主角。他从小被一群外星窃贼和走私犯抚养长大，并在童年时期从地球被绑架。他是半人半天神，拥有挥舞一种名为“无限宝石”的能量武器的能力。他定于 2021 年 5 月回归 MCU。
在 Graph RAG 中包含 nl2graphquery 作为上下文¶
(子)图 RAG 和 nl2graphquery 的性质不同。它们之间没有优劣之分，只是在某些类型的问题中，其中一种更适合。要了解它们之间的更多区别，请参阅此演示比较两者。
在实际应用中，我们可能无法总是知道哪种方法效果更好，因此，在 RAG 中充分利用 KG 的一种方法是同时获取两种检索结果作为上下文，并让 LLM + Prompt 在所有相关信息的基础上生成答案。
因此，作为可选操作，我们可以选择从 KG 中检索到的两部分上下文合成答案
- Graph RAG，默认的检索方法，提取与问题中关键实体相关的子图。
- NL2GraphQuery，根据查询和知识图谱的 Schema 生成知识图谱查询，此功能默认关闭。
我们可以设置 with_nl2graphquery=True 来启用它，如下所示
In [ ]
已复制！
graph_rag_retriever_with_nl2graphquery = KnowledgeGraphRAGRetriever(
storage_context=storage_context,
verbose=True,
with_nl2graphquery=True,
)
query_engine_with_nl2graphquery = RetrieverQueryEngine.from_args(
graph_rag_retriever_with_nl2graphquery,
)
graph_rag_retriever_with_nl2graphquery = KnowledgeGraphRAGRetriever( storage_context=storage_context, verbose=True, with_nl2graphquery=True, ) query_engine_with_nl2graphquery = RetrieverQueryEngine.from_args( graph_rag_retriever_with_nl2graphquery, )
In [ ]
已复制！
response = query_engine_with_nl2graphquery.query(
"What do you know about Peter Quill?",
)
display(Markdown(f"<b>{response}</b>"))
response = query_engine_with_nl2graphquery.query( "What do you know about Peter Quill?", ) display(Markdown(f"{response}"))
Graph Store Query: ``` MATCH (p:`entity`)-[:`relationship`]->(m:`entity`) WHERE p.`entity`.`name` == 'Peter Quill' RETURN m.`entity`.`name`; ``` Graph Store Response: {'m.entity.name': ['May 2021', 'as a child', 'Guardians of the Galaxy', 'a group of alien thieves and smugglers', 'half-Celestial']} Entities processed: ['Star', 'Lord', 'Marvel', 'Quill', 'Galaxy', 'Guardians', 'Guardians of the Galaxy', 'Star-Lord', 'Peter Quill', 'Peter'] Entities processed: ['Star', 'Lord', 'Marvel', 'Quill', 'Galaxy', 'Guardians', 'Guardians of the Galaxy', 'Star-Lord', 'Peter Quill', 'Peter'] Graph RAG context: The following are knowledge sequence in max depth 2 in the form of `subject predicate, object, predicate_next_hop, object_next_hop ...` extracted based on key entities as subject: Guardians, is member of, Guardians, was experimented on, by the High Evolutionary Guardians, is member of, Guardians, considered to tell, origins Guardians, is member of, Guardians, origins, team-up movie Guardians, is member of, Guardians, befriended, his fellow Batch 89 test subjects Guardians, is member of, Guardians, sought to enhance and anthropomorphize animal lifeforms, to create an ideal society Guardians, is member of, Guardians, is creator of, Rocket Guardians, is member of, Guardians, is, Mantis Guardians, is member of, Guardians, is half-sister of, Mantis Guardians, is member of, Guardians, is, Kraglin Guardians, is member of, Guardians, developed psionic abilities, after being abandoned in outer space Guardians, is member of, Guardians, would portray, Cosmo Guardians, is member of, Guardians, recalls, his past Guardians, is member of, Guardians Guardians, is member of, Guardians, focus on, third Guardians-centric film Guardians, is member of, Guardians, is, Rocket Guardians, is member of, Guardians, backstory, flashbacks Guardians, is member of, Guardians, is former second-in-command of, Ravagers Quill, is half-sister of, Mantis, is member of, Guardians Quill, is half-sister of, Mantis, is, Mantis Quill, is in a state of depression, following the appearance of a variant of his dead lover Gamora Quill, is half-sister of, Mantis Peter Quill, is leader of, Guardians of the Galaxy, is sequel to, Guardians of the Galaxy Peter Quill, was raised by, a group of alien thieves and smugglers Peter Quill, would return to the MCU, May 2021 Peter Quill, is leader of, Guardians of the Galaxy Peter Quill, is half-human, half-Celestial Peter Quill, was abducted from Earth, as a child Guardians of the Galaxy, is sequel to, Guardians of the Galaxy, released in, Dolby Cinema Guardians of the Galaxy, is sequel to, Guardians of the Galaxy, released on, Disney+ Guardians of the Galaxy, is sequel to, Guardians of the Galaxy, is sequel to, Guardians of the Galaxy Vol. 2
彼得·奎尔是银河护卫队的领袖，从小从地球被绑架。他是半人半天神，由一群外星窃贼和走私犯抚养长大。他将于 2021 年 5 月回归 MCU。
让我们检查响应的元数据，通过查看 response.metadata 来了解更多关于使用 nl2graphquery 进行 Graph RAG 检索的详细信息。
- text2Cypher，它生成一个针对答案的 Cypher 查询作为上下文。
Graph Store Query: MATCH (e:`entity`)-[r:`relationship`]->(e2:`entity`)
WHERE e.`entity`.`name` == 'Peter Quill'
RETURN e2.`entity`.`name`
SubGraph RAG，它获取“彼得·奎尔”的子图来构建上下文。
最后，它结合了两个上下文节点，合成答案。
In [ ]
已复制！
import pprint
pp = pprint.PrettyPrinter()
pp.pprint(response.metadata)
import pprint pp = pprint.PrettyPrinter() pp.pprint(response.metadata)
{'46faf6d6-8a71-44c8-ae81-794e71a62fbc': {'graph_schema': 'Node properties: '
"[{'tag': 'entity', "
"'properties': "
"[('name', "
"'string')]}]\n"
'Edge properties: '
"[{'edge': "
"'relationship', "
"'properties': "
"[('relationship', "
"'string')]}]\n"
'Relationships: '
"['(:entity)-[:relationship]->(:entity)']\n",
'graph_store_query': '```\n'
'MATCH '
'(p:`entity`)-[:`relationship`]->(m:`entity`) '
'WHERE '
'p.`entity`.`name` '
"== 'Peter "
"Quill'\n"
'RETURN '
'm.`entity`.`name`;\n'
'```',
'graph_store_response': {'m.entity.name': ['May '
'2021',
'as '
'a '
'child',
'Guardians '
'of '
'the '
'Galaxy',
'a '
'group '
'of '
'alien '
'thieves '
'and '
'smugglers',
'half-Celestial']},
'query_str': 'What do you know about '
'Peter Quill?'},
'def19bbf-d8ac-43b2-a121-675748cc9454': {'kg_rel_map': {'Guardians': ['Guardians, '
'is '
'member '
'of, '
'Guardians, '
'was '
'experimented '
'on, by '
'the '
'High '
'Evolutionary',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'considered '
'to '
'tell, '
'origins',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'origins, '
'team-up '
'movie',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'befriended, '
'his '
'fellow '
'Batch '
'89 '
'test '
'subjects',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'sought '
'to '
'enhance '
'and '
'anthropomorphize '
'animal '
'lifeforms, '
'to '
'create '
'an '
'ideal '
'society',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'is '
'creator '
'of, '
'Rocket',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'is, '
'Mantis',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'is '
'half-sister '
'of, '
'Mantis',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'is, '
'Kraglin',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'developed '
'psionic '
'abilities, '
'after '
'being '
'abandoned '
'in '
'outer '
'space',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'would '
'portray, '
'Cosmo',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'recalls, '
'his '
'past',
'Guardians, '
'is '
'member '
'of, '
'Guardians',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'focus '
'on, '
'third '
'Guardians-centric '
'film',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'is, '
'Rocket',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'backstory, '
'flashbacks',
'Guardians, '
'is '
'member '
'of, '
'Guardians, '
'is '
'former '
'second-in-command '
'of, '
'Ravagers'],
'Guardians of the Galaxy': ['Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'in, '
'Dolby '
'Cinema',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'on, '
'Disney+',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy '
'Vol. '
'2',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'in, '
'3D',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'in, '
'4DX',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'grossed, '
'$32 '
'million '
'in '
'its '
'third '
'weekend',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'leader '
'of, '
'Guardians '
'of '
'the '
'Galaxy',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'in, '
'Guardians '
'of '
'the '
'Galaxy '
'Vol. '
'3',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'wrote '
'and '
'directed, '
'Guardians '
'of '
'the '
'Galaxy '
'Vol. '
'3',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'is, '
'American '
'superhero '
'film',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'grossed, '
'$845.4 '
'million',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'was '
'fired '
'from, '
'Guardians '
'of '
'the '
'Galaxy '
'Vol. '
'3',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'leader '
'of, '
'Guardians '
'of '
'the '
'Galaxy, '
'was '
'abducted '
'from '
'Earth, '
'as '
'a '
'child',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'grossed, '
'$359 '
'million '
'in '
'the '
'United '
'States '
'and '
'Canada',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'on, '
'digital '
'download',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'in, '
'IMAX',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'leader '
'of, '
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'half-human, '
'half-Celestial',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'leader '
'of, '
'Guardians '
'of '
'the '
'Galaxy, '
'was '
'raised '
'by, '
'a '
'group '
'of '
'alien '
'thieves '
'and '
'smugglers',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'screened '
'at, '
'Dongdaemun '
'Design '
'Plaza',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'in, '
'ScreenX',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'leader '
'of, '
'Guardians '
'of '
'the '
'Galaxy, '
'would '
'return '
'to '
'the '
'MCU, '
'May '
'2021',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'grossed, '
'$486.4 '
'million '
'in '
'other '
'territories',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'on, '
'Ultra '
'HD '
'Blu-ray',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'on, '
'DVD',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'grossed, '
'$92 '
'million '
'for '
'a '
'drop '
'of '
'40% '
'from '
'its '
'opening '
'weekend',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'premiered '
'at, '
'Disneyland '
'Paris',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'released '
'on, '
'Blu-ray',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'could '
'happen, '
'April '
'2017',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'made, '
'$48.2 '
'million '
'on '
'its '
'first '
'day',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'grossed, '
'$168.1 '
'million '
'in '
'its '
'opening '
'weekend',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'debuted '
'with, '
'$118.4 '
'million',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'would '
'likely '
'center '
'on, '
'new '
'group '
'of '
'characters',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'retained '
'the '
'top '
'spot '
'at '
'the '
'box '
'office '
'with, '
'$62 '
'million',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'be '
'his '
'last '
'Guardians '
'film, '
'September '
'2019',
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy, '
'nominated '
'for, '
'Best '
'Picture'],
'Marvel': ['Marvel, '
'was fired '
'from, '
'Marvel, '
'stated, '
'that in '
'addition '
'to having '
'the '
'basic '
'story '
'for '
'Guardians '
'of the '
'Galaxy '
'Vol.2 '
'(2017) '
'while '
'working '
'on the '
'first '
'film',
'Marvel, '
'was fired '
'from, '
'Marvel, '
'was '
'unsure, '
'if he '
'would be '
'involved '
'with a '
'third '
'Guardians '
'film',
'Marvel, '
'was fired '
'from, '
'Marvel, '
'was '
'privately '
'notified '
'by, Horn',
'Marvel, '
'was fired '
'from, '
'Marvel, '
'was fired '
'from, '
'Guardians '
'of the '
'Galaxy '
'Vol. 3',
'Marvel, '
'was fired '
'from, '
'Marvel',
'Marvel, '
'was fired '
'from, '
'Marvel, '
'wrote and '
'directed, '
'Guardians '
'of the '
'Galaxy '
'Vol. 3',
'Marvel, '
'was fired '
'from, '
'Marvel, '
'was fired '
'from, '
'Disney',
'Marvel, '
'was fired '
'from, '
'Marvel, '
'could '
'return as '
'director '
'for, '
'Vol.3'],
'Peter Quill': ['Peter '
'Quill, '
'is '
'leader '
'of, '
'Guardians '
'of '
'the '
'Galaxy, '
'is '
'sequel '
'to, '
'Guardians '
'of '
'the '
'Galaxy',
'Peter '
'Quill, '
'was '
'raised '
'by, '
'a '
'group '
'of '
'alien '
'thieves '
'and '
'smugglers',
'Peter '
'Quill, '
'would '
'return '
'to '
'the '
'MCU, '
'May '
'2021',
'Peter '
'Quill, '
'is '
'leader '
'of, '
'Guardians '
'of '
'the '
'Galaxy',
'Peter '
'Quill, '
'is '
'half-human, '
'half-Celestial',
'Peter '
'Quill, '
'was '
'abducted '
'from '
'Earth, '
'as a '
'child'],
'Quill': ['Quill, is '
'half-sister '
'of, '
'Mantis, is '
'member of, '
'Guardians',
'Quill, is '
'half-sister '
'of, '
'Mantis, '
'is, Mantis',
'Quill, is '
'in a state '
'of '
'depression, '
'following '
'the '
'appearance '
'of a '
'variant of '
'his dead '
'lover '
'Gamora',
'Quill, is '
'half-sister '
'of, '
'Mantis']},
'kg_rel_text': ['Guardians, is '
'member of, '
'Guardians, was '
'experimented on, by '
'the High '
'Evolutionary',
'Guardians, is '
'member of, '
'Guardians, '
'considered to tell, '
'origins',
'Guardians, is '
'member of, '
'Guardians, origins, '
'team-up movie',
'Guardians, is '
'member of, '
'Guardians, '
'befriended, his '
'fellow Batch 89 '
'test subjects',
'Guardians, is '
'member of, '
'Guardians, sought '
'to enhance and '
'anthropomorphize '
'animal lifeforms, '
'to create an ideal '
'society',
'Guardians, is '
'member of, '
'Guardians, is '
'creator of, Rocket',
'Guardians, is '
'member of, '
'Guardians, is, '
'Mantis',
'Guardians, is '
'member of, '
'Guardians, is '
'half-sister of, '
'Mantis',
'Guardians, is '
'member of, '
'Guardians, is, '
'Kraglin',
'Guardians, is '
'member of, '
'Guardians, '
'developed psionic '
'abilities, after '
'being abandoned in '
'outer space',
'Guardians, is '
'member of, '
'Guardians, would '
'portray, Cosmo',
'Guardians, is '
'member of, '
'Guardians, recalls, '
'his past',
'Guardians, is '
'member of, '
'Guardians',
'Guardians, is '
'member of, '
'Guardians, focus '
'on, third '
'Guardians-centric '
'film',
'Guardians, is '
'member of, '
'Guardians, is, '
'Rocket',
'Guardians, is '
'member of, '
'Guardians, '
'backstory, '
'flashbacks',
'Guardians, is '
'member of, '
'Guardians, is '
'former '
'second-in-command '
'of, Ravagers',
'Quill, is '
'half-sister of, '
'Mantis, is member '
'of, Guardians',
'Quill, is '
'half-sister of, '
'Mantis, is, Mantis',
'Quill, is in a '
'state of '
'depression, '
'following the '
'appearance of a '
'variant of his dead '
'lover Gamora',
'Quill, is '
'half-sister of, '
'Mantis',
'Peter Quill, is '
'leader of, '
'Guardians of the '
'Galaxy, is sequel '
'to, Guardians of '
'the Galaxy',
'Peter Quill, was '
'raised by, a group '
'of alien thieves '
'and smugglers',
'Peter Quill, would '
'return to the MCU, '
'May 2021',
'Peter Quill, is '
'leader of, '
'Guardians of the '
'Galaxy',
'Peter Quill, is '
'half-human, '
'half-Celestial',
'Peter Quill, was '
'abducted from '
'Earth, as a child',
'Guardians of the '
'Galaxy, is sequel '
'to, Guardians of '
'the Galaxy, '
'released in, Dolby '
'Cinema',
'Guardians of the '
'Galaxy, is sequel '
'to, Guardians of '
'the Galaxy, '
'released on, '
'Disney+',
'Guardians of the '
'Galaxy, is sequel '
'to, Guardians of '
'the Galaxy, is '
'sequel to, '
'Guardians of the '
'Galaxy Vol. 2']}}
