Global Search

In [1]:

# Copyright (c) 2024 Microsoft Corporation.
# Licensed under the MIT License.

# Copyright (c) 2024 Microsoft Corporation. # Licensed under the MIT License.

In [2]:

import os

import pandas as pd
import tiktoken

from graphrag.query.indexer_adapters import (
    read_indexer_communities,
    read_indexer_entities,
    read_indexer_reports,
)
from graphrag.query.llm.oai.chat_openai import ChatOpenAI
from graphrag.query.llm.oai.typing import OpenaiApiType
from graphrag.query.structured_search.global_search.community_context import (
    GlobalCommunityContext,
)
from graphrag.query.structured_search.global_search.search import GlobalSearch

import os import pandas as pd import tiktoken from graphrag.query.indexer_adapters import ( read_indexer_communities, read_indexer_entities, read_indexer_reports, ) from graphrag.query.llm.oai.chat_openai import ChatOpenAI from graphrag.query.llm.oai.typing import OpenaiApiType from graphrag.query.structured_search.global_search.community_context import ( GlobalCommunityContext, ) from graphrag.query.structured_search.global_search.search import GlobalSearch

Global Search example¶

Global search method generates answers by searching over all AI-generated community reports in a map-reduce fashion. This is a resource-intensive method, but often gives good responses for questions that require an understanding of the dataset as a whole (e.g. What are the most significant values of the herbs mentioned in this notebook?).

LLM setup¶

In [3]:

api_key = os.environ["GRAPHRAG_API_KEY"]
llm_model = os.environ["GRAPHRAG_LLM_MODEL"]

llm = ChatOpenAI(
    api_key=api_key,
    model=llm_model,
    api_type=OpenaiApiType.OpenAI,  # OpenaiApiType.OpenAI or OpenaiApiType.AzureOpenAI
    max_retries=20,
)

token_encoder = tiktoken.encoding_for_model(llm_model)

api_key = os.environ["GRAPHRAG_API_KEY"] llm_model = os.environ["GRAPHRAG_LLM_MODEL"] llm = ChatOpenAI( api_key=api_key, model=llm_model, api_type=OpenaiApiType.OpenAI, # OpenaiApiType.OpenAI or OpenaiApiType.AzureOpenAI max_retries=20, ) token_encoder = tiktoken.encoding_for_model(llm_model)

Load community reports as context for global search¶

• Load all community reports in the create_final_community_reports table from the GraphRAG, to be used as context data for global search.
• Load entities from the create_final_nodes and create_final_entities tables from the GraphRAG, to be used for calculating community weights for context ranking. Note that this is optional (if no entities are provided, we will not calculate community weights and only use the rank attribute in the community reports table for context ranking)
• Load all communities in the create_final_communites table from the GraphRAG, to be used to reconstruct the community graph hierarchy for dynamic community selection.

In [4]:

# parquet files generated from indexing pipeline
INPUT_DIR = "./inputs/operation dulce"
COMMUNITY_TABLE = "create_final_communities"
COMMUNITY_REPORT_TABLE = "create_final_community_reports"
ENTITY_TABLE = "create_final_nodes"
ENTITY_EMBEDDING_TABLE = "create_final_entities"

# community level in the Leiden community hierarchy from which we will load the community reports
# higher value means we use reports from more fine-grained communities (at the cost of higher computation cost)
COMMUNITY_LEVEL = 2

# parquet files generated from indexing pipeline INPUT_DIR = "./inputs/operation dulce" COMMUNITY_TABLE = "create_final_communities" COMMUNITY_REPORT_TABLE = "create_final_community_reports" ENTITY_TABLE = "create_final_nodes" ENTITY_EMBEDDING_TABLE = "create_final_entities" # community level in the Leiden community hierarchy from which we will load the community reports # higher value means we use reports from more fine-grained communities (at the cost of higher computation cost) COMMUNITY_LEVEL = 2

In [5]:

community_df = pd.read_parquet(f"{INPUT_DIR}/{COMMUNITY_TABLE}.parquet")
entity_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_TABLE}.parquet")
report_df = pd.read_parquet(f"{INPUT_DIR}/{COMMUNITY_REPORT_TABLE}.parquet")
entity_embedding_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_EMBEDDING_TABLE}.parquet")

communities = read_indexer_communities(community_df, entity_df, report_df)
reports = read_indexer_reports(report_df, entity_df, COMMUNITY_LEVEL)
entities = read_indexer_entities(entity_df, entity_embedding_df, COMMUNITY_LEVEL)

print(f"Total report count: {len(report_df)}")
print(
    f"Report count after filtering by community level {COMMUNITY_LEVEL}: {len(reports)}"
)

report_df.head()

community_df = pd.read_parquet(f"{INPUT_DIR}/{COMMUNITY_TABLE}.parquet") entity_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_TABLE}.parquet") report_df = pd.read_parquet(f"{INPUT_DIR}/{COMMUNITY_REPORT_TABLE}.parquet") entity_embedding_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_EMBEDDING_TABLE}.parquet") communities = read_indexer_communities(community_df, entity_df, report_df) reports = read_indexer_reports(report_df, entity_df, COMMUNITY_LEVEL) entities = read_indexer_entities(entity_df, entity_embedding_df, COMMUNITY_LEVEL) print(f"Total report count: {len(report_df)}") print( f"Report count after filtering by community level {COMMUNITY_LEVEL}: {len(reports)}" ) report_df.head()

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
/tmp/ipykernel_1934/1512985616.py in ?()
      2 entity_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_TABLE}.parquet")
      3 report_df = pd.read_parquet(f"{INPUT_DIR}/{COMMUNITY_REPORT_TABLE}.parquet")
      4 entity_embedding_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_EMBEDDING_TABLE}.parquet")
      5 
----> 6 communities = read_indexer_communities(community_df, entity_df, report_df)
      7 reports = read_indexer_reports(report_df, entity_df, COMMUNITY_LEVEL)
      8 entities = read_indexer_entities(entity_df, entity_embedding_df, COMMUNITY_LEVEL)
      9 

~/work/graphrag/graphrag/graphrag/query/indexer_adapters.py in ?(final_communities, final_nodes, final_community_reports)
    180     reports_df = final_community_reports
    181 
    182     # ensure communities matches community reports
    183     missing_reports = communities_df[
--> 184         ~communities_df.community.isin(reports_df.community.unique())
    185     ].community.to_list()
    186     if len(missing_reports):
    187         log.warning("Missing reports for communities: %s", missing_reports)

~/.cache/pypoetry/virtualenvs/graphrag-F2jvqev7-py3.11/lib/python3.11/site-packages/pandas/core/generic.py in ?(self, name)
   6295             and name not in self._accessors
   6296             and self._info_axis._can_hold_identifiers_and_holds_name(name)
   6297         ):
   6298             return self[name]
-> 6299         return object.__getattribute__(self, name)

AttributeError: 'DataFrame' object has no attribute 'community'

Build global context based on community reports¶

In [6]:

context_builder = GlobalCommunityContext(
    community_reports=reports,
    communities=communities,
    entities=entities,  # default to None if you don't want to use community weights for ranking
    token_encoder=token_encoder,
)

context_builder = GlobalCommunityContext( community_reports=reports, communities=communities, entities=entities, # default to None if you don't want to use community weights for ranking token_encoder=token_encoder, )

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[6], line 2
      1 context_builder = GlobalCommunityContext(
----> 2     community_reports=reports,
      3     communities=communities,
      4     entities=entities,  # default to None if you don't want to use community weights for ranking
      5     token_encoder=token_encoder,
      6 )

NameError: name 'reports' is not defined

Perform global search¶

In [7]:

context_builder_params = {
    "use_community_summary": False,  # False means using full community reports. True means using community short summaries.
    "shuffle_data": True,
    "include_community_rank": True,
    "min_community_rank": 0,
    "community_rank_name": "rank",
    "include_community_weight": True,
    "community_weight_name": "occurrence weight",
    "normalize_community_weight": True,
    "max_tokens": 12_000,  # change this based on the token limit you have on your model (if you are using a model with 8k limit, a good setting could be 5000)
    "context_name": "Reports",
}

map_llm_params = {
    "max_tokens": 1000,
    "temperature": 0.0,
    "response_format": {"type": "json_object"},
}

reduce_llm_params = {
    "max_tokens": 2000,  # change this based on the token limit you have on your model (if you are using a model with 8k limit, a good setting could be 1000-1500)
    "temperature": 0.0,
}

context_builder_params = { "use_community_summary": False, # False means using full community reports. True means using community short summaries. "shuffle_data": True, "include_community_rank": True, "min_community_rank": 0, "community_rank_name": "rank", "include_community_weight": True, "community_weight_name": "occurrence weight", "normalize_community_weight": True, "max_tokens": 12_000, # change this based on the token limit you have on your model (if you are using a model with 8k limit, a good setting could be 5000) "context_name": "Reports", } map_llm_params = { "max_tokens": 1000, "temperature": 0.0, "response_format": {"type": "json_object"}, } reduce_llm_params = { "max_tokens": 2000, # change this based on the token limit you have on your model (if you are using a model with 8k limit, a good setting could be 1000-1500) "temperature": 0.0, }

In [8]:

search_engine = GlobalSearch(
    llm=llm,
    context_builder=context_builder,
    token_encoder=token_encoder,
    max_data_tokens=12_000,  # change this based on the token limit you have on your model (if you are using a model with 8k limit, a good setting could be 5000)
    map_llm_params=map_llm_params,
    reduce_llm_params=reduce_llm_params,
    allow_general_knowledge=False,  # set this to True will add instruction to encourage the LLM to incorporate general knowledge in the response, which may increase hallucinations, but could be useful in some use cases.
    json_mode=True,  # set this to False if your LLM model does not support JSON mode.
    context_builder_params=context_builder_params,
    concurrent_coroutines=32,
    response_type="multiple paragraphs",  # free form text describing the response type and format, can be anything, e.g. prioritized list, single paragraph, multiple paragraphs, multiple-page report
)

search_engine = GlobalSearch( llm=llm, context_builder=context_builder, token_encoder=token_encoder, max_data_tokens=12_000, # change this based on the token limit you have on your model (if you are using a model with 8k limit, a good setting could be 5000) map_llm_params=map_llm_params, reduce_llm_params=reduce_llm_params, allow_general_knowledge=False, # set this to True will add instruction to encourage the LLM to incorporate general knowledge in the response, which may increase hallucinations, but could be useful in some use cases. json_mode=True, # set this to False if your LLM model does not support JSON mode. context_builder_params=context_builder_params, concurrent_coroutines=32, response_type="multiple paragraphs", # free form text describing the response type and format, can be anything, e.g. prioritized list, single paragraph, multiple paragraphs, multiple-page report )

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[8], line 3
      1 search_engine = GlobalSearch(
      2     llm=llm,
----> 3     context_builder=context_builder,
      4     token_encoder=token_encoder,
      5     max_data_tokens=12_000,  # change this based on the token limit you have on your model (if you are using a model with 8k limit, a good setting could be 5000)
      6     map_llm_params=map_llm_params,
      7     reduce_llm_params=reduce_llm_params,
      8     allow_general_knowledge=False,  # set this to True will add instruction to encourage the LLM to incorporate general knowledge in the response, which may increase hallucinations, but could be useful in some use cases.
      9     json_mode=True,  # set this to False if your LLM model does not support JSON mode.
     10     context_builder_params=context_builder_params,
     11     concurrent_coroutines=32,
     12     response_type="multiple paragraphs",  # free form text describing the response type and format, can be anything, e.g. prioritized list, single paragraph, multiple paragraphs, multiple-page report
     13 )

NameError: name 'context_builder' is not defined

In [9]:

result = await search_engine.asearch(
    "What is Cosmic Vocalization and who are involved in it?"
)

print(result.response)

result = await search_engine.asearch( "What is Cosmic Vocalization and who are involved in it?" ) print(result.response)

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[9], line 1
----> 1 result = await search_engine.asearch(
      2     "What is Cosmic Vocalization and who are involved in it?"
      3 )
      5 print(result.response)

NameError: name 'search_engine' is not defined

In [10]:

# inspect the data used to build the context for the LLM responses
result.context_data["reports"]

# inspect the data used to build the context for the LLM responses result.context_data["reports"]

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[10], line 2
      1 # inspect the data used to build the context for the LLM responses
----> 2 result.context_data["reports"]

NameError: name 'result' is not defined

In [11]:

# inspect number of LLM calls and tokens
print(
    f"LLM calls: {result.llm_calls}. Prompt tokens: {result.prompt_tokens}. Output tokens: {result.output_tokens}."
)

# inspect number of LLM calls and tokens print( f"LLM calls: {result.llm_calls}. Prompt tokens: {result.prompt_tokens}. Output tokens: {result.output_tokens}." )

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[11], line 3
      1 # inspect number of LLM calls and tokens
      2 print(
----> 3     f"LLM calls: {result.llm_calls}. Prompt tokens: {result.prompt_tokens}. Output tokens: {result.output_tokens}."
      4 )

NameError: name 'result' is not defined