Local 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.context_builder.entity_extraction import EntityVectorStoreKey
from graphrag.query.indexer_adapters import (
    read_indexer_covariates,
    read_indexer_entities,
    read_indexer_relationships,
    read_indexer_reports,
    read_indexer_text_units,
)
from graphrag.query.llm.oai.chat_openai import ChatOpenAI
from graphrag.query.llm.oai.embedding import OpenAIEmbedding
from graphrag.query.llm.oai.typing import OpenaiApiType
from graphrag.query.question_gen.local_gen import LocalQuestionGen
from graphrag.query.structured_search.local_search.mixed_context import (
    LocalSearchMixedContext,
)
from graphrag.query.structured_search.local_search.search import LocalSearch
from graphrag.vector_stores.lancedb import LanceDBVectorStore

import os import pandas as pd import tiktoken from graphrag.query.context_builder.entity_extraction import EntityVectorStoreKey from graphrag.query.indexer_adapters import ( read_indexer_covariates, read_indexer_entities, read_indexer_relationships, read_indexer_reports, read_indexer_text_units, ) from graphrag.query.llm.oai.chat_openai import ChatOpenAI from graphrag.query.llm.oai.embedding import OpenAIEmbedding from graphrag.query.llm.oai.typing import OpenaiApiType from graphrag.query.question_gen.local_gen import LocalQuestionGen from graphrag.query.structured_search.local_search.mixed_context import ( LocalSearchMixedContext, ) from graphrag.query.structured_search.local_search.search import LocalSearch from graphrag.vector_stores.lancedb import LanceDBVectorStore

Local Search Example¶

Local search method generates answers by combining relevant data from the AI-extracted knowledge-graph with text chunks of the raw documents. This method is suitable for questions that require an understanding of specific entities mentioned in the documents (e.g. What are the healing properties of chamomile?).

Load text units and graph data tables as context for local search¶

• In this test we first load indexing outputs from parquet files to dataframes, then convert these dataframes into collections of data objects aligning with the knowledge model.

Load tables to dataframes¶

In [3]:

INPUT_DIR = "./inputs/operation dulce"
LANCEDB_URI = f"{INPUT_DIR}/lancedb"

COMMUNITY_REPORT_TABLE = "create_final_community_reports"
ENTITY_TABLE = "create_final_nodes"
ENTITY_EMBEDDING_TABLE = "create_final_entities"
RELATIONSHIP_TABLE = "create_final_relationships"
COVARIATE_TABLE = "create_final_covariates"
TEXT_UNIT_TABLE = "create_final_text_units"
COMMUNITY_LEVEL = 2

INPUT_DIR = "./inputs/operation dulce" LANCEDB_URI = f"{INPUT_DIR}/lancedb" COMMUNITY_REPORT_TABLE = "create_final_community_reports" ENTITY_TABLE = "create_final_nodes" ENTITY_EMBEDDING_TABLE = "create_final_entities" RELATIONSHIP_TABLE = "create_final_relationships" COVARIATE_TABLE = "create_final_covariates" TEXT_UNIT_TABLE = "create_final_text_units" COMMUNITY_LEVEL = 2

Read entities¶

In [4]:

# read nodes table to get community and degree data
entity_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_TABLE}.parquet")
entity_embedding_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_EMBEDDING_TABLE}.parquet")

entities = read_indexer_entities(entity_df, entity_embedding_df, COMMUNITY_LEVEL)

# load description embeddings to an in-memory lancedb vectorstore
# to connect to a remote db, specify url and port values.
description_embedding_store = LanceDBVectorStore(
    collection_name="default-entity-description",
)
description_embedding_store.connect(db_uri=LANCEDB_URI)

print(f"Entity count: {len(entity_df)}")
entity_df.head()

# read nodes table to get community and degree data entity_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_TABLE}.parquet") entity_embedding_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_EMBEDDING_TABLE}.parquet") entities = read_indexer_entities(entity_df, entity_embedding_df, COMMUNITY_LEVEL) # load description embeddings to an in-memory lancedb vectorstore # to connect to a remote db, specify url and port values. description_embedding_store = LanceDBVectorStore( collection_name="default-entity-description", ) description_embedding_store.connect(db_uri=LANCEDB_URI) print(f"Entity count: {len(entity_df)}") entity_df.head()

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[4], line 5
      2 entity_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_TABLE}.parquet")
      3 entity_embedding_df = pd.read_parquet(f"{INPUT_DIR}/{ENTITY_EMBEDDING_TABLE}.parquet")
----> 5 entities = read_indexer_entities(entity_df, entity_embedding_df, COMMUNITY_LEVEL)
      7 # load description embeddings to an in-memory lancedb vectorstore
      8 # to connect to a remote db, specify url and port values.
      9 description_embedding_store = LanceDBVectorStore(
     10     collection_name="default-entity-description",
     11 )

File ~/work/graphrag/graphrag/graphrag/query/indexer_adapters.py:154, in read_indexer_entities(final_nodes, final_entities, community_level)
    149 final_df = nodes_df.merge(entities_df, on="id", how="inner").drop_duplicates(
    150     subset=["id"]
    151 )
    153 # read entity dataframe to knowledge model objects
--> 154 return read_entities(
    155     df=final_df,
    156     id_col="id",
    157     title_col="title",
    158     type_col="type",
    159     short_id_col="human_readable_id",
    160     description_col="description",
    161     community_col="community",
    162     rank_col="degree",
    163     name_embedding_col=None,
    164     description_embedding_col="description_embedding",
    165     text_unit_ids_col="text_unit_ids",
    166 )

File ~/work/graphrag/graphrag/graphrag/query/input/loaders/dfs.py:44, in read_entities(df, id_col, short_id_col, title_col, type_col, description_col, name_embedding_col, description_embedding_col, community_col, text_unit_ids_col, rank_col, attributes_cols)
     39 entities = []
     40 for idx, row in df.iterrows():
     41     entity = Entity(
     42         id=to_str(row, id_col),
     43         short_id=to_optional_str(row, short_id_col) if short_id_col else str(idx),
---> 44         title=to_str(row, title_col),
     45         type=to_optional_str(row, type_col),
     46         description=to_optional_str(row, description_col),
     47         name_embedding=to_optional_list(row, name_embedding_col, item_type=float),
     48         description_embedding=to_optional_list(
     49             row, description_embedding_col, item_type=float
     50         ),
     51         community_ids=to_optional_list(row, community_col, item_type=str),
     52         text_unit_ids=to_optional_list(row, text_unit_ids_col),
     53         rank=to_optional_int(row, rank_col),
     54         attributes=(
     55             {col: row.get(col) for col in attributes_cols}
     56             if attributes_cols
     57             else None
     58         ),
     59     )
     60     entities.append(entity)
     61 return entities

File ~/work/graphrag/graphrag/graphrag/query/input/loaders/utils.py:19, in to_str(data, column_name)
     17     return str(data[column_name])
     18 msg = f"Column {column_name} not found in data"
---> 19 raise ValueError(msg)

ValueError: Column title not found in data

Read relationships¶

In [5]:

relationship_df = pd.read_parquet(f"{INPUT_DIR}/{RELATIONSHIP_TABLE}.parquet")
relationships = read_indexer_relationships(relationship_df)

print(f"Relationship count: {len(relationship_df)}")
relationship_df.head()

relationship_df = pd.read_parquet(f"{INPUT_DIR}/{RELATIONSHIP_TABLE}.parquet") relationships = read_indexer_relationships(relationship_df) print(f"Relationship count: {len(relationship_df)}") relationship_df.head()

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[5], line 2
      1 relationship_df = pd.read_parquet(f"{INPUT_DIR}/{RELATIONSHIP_TABLE}.parquet")
----> 2 relationships = read_indexer_relationships(relationship_df)
      4 print(f"Relationship count: {len(relationship_df)}")
      5 relationship_df.head()

File ~/work/graphrag/graphrag/graphrag/query/indexer_adapters.py:66, in read_indexer_relationships(final_relationships)
     64 def read_indexer_relationships(final_relationships: pd.DataFrame) -> list[Relationship]:
     65     """Read in the Relationships from the raw indexing outputs."""
---> 66     return read_relationships(
     67         df=final_relationships,
     68         short_id_col="human_readable_id",
     69         rank_col="combined_degree",
     70         description_embedding_col=None,
     71         attributes_cols=None,
     72     )

File ~/work/graphrag/graphrag/graphrag/query/input/loaders/dfs.py:91, in read_relationships(df, id_col, short_id_col, source_col, target_col, description_col, rank_col, description_embedding_col, weight_col, text_unit_ids_col, attributes_cols)
     78 relationships = []
     79 for idx, row in df.iterrows():
     80     rel = Relationship(
     81         id=to_str(row, id_col),
     82         short_id=to_optional_str(row, short_id_col) if short_id_col else str(idx),
     83         source=to_str(row, source_col),
     84         target=to_str(row, target_col),
     85         description=to_optional_str(row, description_col),
     86         description_embedding=to_optional_list(
     87             row, description_embedding_col, item_type=float
     88         ),
     89         weight=to_optional_float(row, weight_col),
     90         text_unit_ids=to_optional_list(row, text_unit_ids_col, item_type=str),
---> 91         rank=to_optional_int(row, rank_col),
     92         attributes=(
     93             {col: row.get(col) for col in attributes_cols}
     94             if attributes_cols
     95             else None
     96         ),
     97     )
     98     relationships.append(rel)
     99 return relationships

File ~/work/graphrag/graphrag/graphrag/query/input/loaders/utils.py:135, in to_optional_int(data, column_name)
    133 else:
    134     msg = f"Column {column_name} not found in data"
--> 135     raise ValueError(msg)
    137 return int(value)

ValueError: Column combined_degree not found in data

In [6]:

# NOTE: covariates are turned off by default, because they generally need prompt tuning to be valuable
# Please see the GRAPHRAG_CLAIM_* settings
covariate_df = pd.read_parquet(f"{INPUT_DIR}/{COVARIATE_TABLE}.parquet")

claims = read_indexer_covariates(covariate_df)

print(f"Claim records: {len(claims)}")
covariates = {"claims": claims}

# NOTE: covariates are turned off by default, because they generally need prompt tuning to be valuable # Please see the GRAPHRAG_CLAIM_* settings covariate_df = pd.read_parquet(f"{INPUT_DIR}/{COVARIATE_TABLE}.parquet") claims = read_indexer_covariates(covariate_df) print(f"Claim records: {len(claims)}") covariates = {"claims": claims}

Claim records: 156

Read community reports¶

In [7]:

report_df = pd.read_parquet(f"{INPUT_DIR}/{COMMUNITY_REPORT_TABLE}.parquet")
reports = read_indexer_reports(report_df, entity_df, COMMUNITY_LEVEL)

print(f"Report records: {len(report_df)}")
report_df.head()

report_df = pd.read_parquet(f"{INPUT_DIR}/{COMMUNITY_REPORT_TABLE}.parquet") reports = read_indexer_reports(report_df, entity_df, COMMUNITY_LEVEL) print(f"Report records: {len(report_df)}") report_df.head()

Report records: 20

Out[7]:

	community	full_content	level	rank	title	rank_explanation	summary	findings	full_content_json	id
0	10	# Paranormal Military Squad at Dulce Base: Dec...	1	8.5	Paranormal Military Squad at Dulce Base: Decod...	The impact severity rating is high due to the ...	The Paranormal Military Squad, stationed at Du...	[{'explanation': 'Jordan is a central figure i...	{\n "title": "Paranormal Military Squad at ...	1ba2d200-dd26-4693-affe-a5539d0a0e0d
1	11	# Dulce and Paranormal Military Squad Operatio...	1	8.5	Dulce and Paranormal Military Squad Operations	The impact severity rating is high due to the ...	The community centers around Dulce, a secretiv...	[{'explanation': 'Dulce is described as a top-...	{\n "title": "Dulce and Paranormal Military...	a8a530b0-ae6b-44ea-b11c-9f70d138298d
2	12	# Paranormal Military Squad and Dulce Base Ope...	1	7.5	Paranormal Military Squad and Dulce Base Opera...	The impact severity rating is relatively high ...	The community centers around the Paranormal Mi...	[{'explanation': 'Taylor is a central figure w...	{\n "title": "Paranormal Military Squad and...	0478975b-c805-4cc1-b746-82f3e689e2f3
3	13	# Mission Dynamics and Leadership: Cruz and Wa...	1	7.5	Mission Dynamics and Leadership: Cruz and Wash...	The impact severity rating is relatively high ...	This report explores the intricate dynamics of...	[{'explanation': 'Cruz is a central figure in ...	{\n "title": "Mission Dynamics and Leadersh...	b56f6e68-3951-4f07-8760-63700944a375
4	14	# Dulce Base and Paranormal Military Squad: Br...	1	8.5	Dulce Base and Paranormal Military Squad: Brid...	The impact severity rating is high due to the ...	The community centers around the Dulce Base, a...	[{'explanation': 'Sam Rivera, a member of the ...	{\n "title": "Dulce Base and Paranormal Mil...	736e7006-d050-4abb-a122-00febf3f540f

Read text units¶

In [8]:

text_unit_df = pd.read_parquet(f"{INPUT_DIR}/{TEXT_UNIT_TABLE}.parquet")
text_units = read_indexer_text_units(text_unit_df)

print(f"Text unit records: {len(text_unit_df)}")
text_unit_df.head()

text_unit_df = pd.read_parquet(f"{INPUT_DIR}/{TEXT_UNIT_TABLE}.parquet") text_units = read_indexer_text_units(text_unit_df) print(f"Text unit records: {len(text_unit_df)}") text_unit_df.head()

Text unit records: 50

Out[8]:

	id	text	n_tokens	document_ids	entity_ids	relationship_ids	covariate_ids
0	06bdff339c02ab35c80fa49320d5da66	# Operation: Dulce\n\n## Chapter 1\n\nThe thru...	600	[958fdd043f17ade63cb13570b59df295]	[b45241d70f0e43fca764df95b2b81f77, 4119fd06010...	[148fffeb994541b2b4b6dcefda7001a8, 89c08e79329...	[439081a3-bfeb-4693-968a-0f6189d8fa50, 3399e3d...
1	28dc4aa41a3e99deb7f354682f3e9904	's authoritarian performance. _Protocols_, Jor...	600	[958fdd043f17ade63cb13570b59df295]	[4119fd06010c494caa07f439b333f4c5, 077d2820ae1...	[9a6f414210e14841a5b0e661aedc898d, db541b72609...	[b09594cb-d4b7-4de4-a1af-97778300eb1b, f7c5ea4...
2	813db3138ef511c34be86f841f68aa8f	, rehearsing the speech for their subordinates...	600	[958fdd043f17ade63cb13570b59df295]	[b45241d70f0e43fca764df95b2b81f77, 077d2820ae1...	[0467928aa65e4a4fba62bdb1467e3a54, 7cc3356d38d...	[7642fe21-7cb9-4428-848b-d2e3f5ab10ca, 6fe3d6c...
3	d0afd106abf3a2966ff88a99eef710db	, weighing his words carefully. "Our tech is t...	600	[958fdd043f17ade63cb13570b59df295]	[b45241d70f0e43fca764df95b2b81f77, 4119fd06010...	[148fffeb994541b2b4b6dcefda7001a8, 89c08e79329...	[956e4c22-e343-4b5d-ad3c-d44ca3ce5fb5]
4	f4c7c95d7bf621c6eb73d331cbf0b608	stepping into their exchange. The authority i...	600	[958fdd043f17ade63cb13570b59df295]	[077d2820ae1845bcbb1803379a3d1eae, 1fd3fa8bb5a...	[478e4c72d8fb46dd8cc9f0691c9878fd, 82b0446e7c9...	[a54eda4d-b5bf-471d-989f-370eb9aff961, 9f85274...

In [9]:

api_key = os.environ["GRAPHRAG_API_KEY"]
llm_model = os.environ["GRAPHRAG_LLM_MODEL"]
embedding_model = os.environ["GRAPHRAG_EMBEDDING_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.get_encoding("cl100k_base")

text_embedder = OpenAIEmbedding(
    api_key=api_key,
    api_base=None,
    api_type=OpenaiApiType.OpenAI,
    model=embedding_model,
    deployment_name=embedding_model,
    max_retries=20,
)

api_key = os.environ["GRAPHRAG_API_KEY"] llm_model = os.environ["GRAPHRAG_LLM_MODEL"] embedding_model = os.environ["GRAPHRAG_EMBEDDING_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.get_encoding("cl100k_base") text_embedder = OpenAIEmbedding( api_key=api_key, api_base=None, api_type=OpenaiApiType.OpenAI, model=embedding_model, deployment_name=embedding_model, max_retries=20, )

Create local search context builder¶

In [10]:

context_builder = LocalSearchMixedContext(
    community_reports=reports,
    text_units=text_units,
    entities=entities,
    relationships=relationships,
    # if you did not run covariates during indexing, set this to None
    covariates=covariates,
    entity_text_embeddings=description_embedding_store,
    embedding_vectorstore_key=EntityVectorStoreKey.ID,  # if the vectorstore uses entity title as ids, set this to EntityVectorStoreKey.TITLE
    text_embedder=text_embedder,
    token_encoder=token_encoder,
)

context_builder = LocalSearchMixedContext( community_reports=reports, text_units=text_units, entities=entities, relationships=relationships, # if you did not run covariates during indexing, set this to None covariates=covariates, entity_text_embeddings=description_embedding_store, embedding_vectorstore_key=EntityVectorStoreKey.ID, # if the vectorstore uses entity title as ids, set this to EntityVectorStoreKey.TITLE text_embedder=text_embedder, token_encoder=token_encoder, )

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[10], line 4
      1 context_builder = LocalSearchMixedContext(
      2     community_reports=reports,
      3     text_units=text_units,
----> 4     entities=entities,
      5     relationships=relationships,
      6     # if you did not run covariates during indexing, set this to None
      7     covariates=covariates,
      8     entity_text_embeddings=description_embedding_store,
      9     embedding_vectorstore_key=EntityVectorStoreKey.ID,  # if the vectorstore uses entity title as ids, set this to EntityVectorStoreKey.TITLE
     10     text_embedder=text_embedder,
     11     token_encoder=token_encoder,
     12 )

NameError: name 'entities' is not defined

Create local search engine¶

In [11]:

# text_unit_prop: proportion of context window dedicated to related text units
# community_prop: proportion of context window dedicated to community reports.
# The remaining proportion is dedicated to entities and relationships. Sum of text_unit_prop and community_prop should be <= 1
# conversation_history_max_turns: maximum number of turns to include in the conversation history.
# conversation_history_user_turns_only: if True, only include user queries in the conversation history.
# top_k_mapped_entities: number of related entities to retrieve from the entity description embedding store.
# top_k_relationships: control the number of out-of-network relationships to pull into the context window.
# include_entity_rank: if True, include the entity rank in the entity table in the context window. Default entity rank = node degree.
# include_relationship_weight: if True, include the relationship weight in the context window.
# include_community_rank: if True, include the community rank in the context window.
# return_candidate_context: if True, return a set of dataframes containing all candidate entity/relationship/covariate records that
# could be relevant. Note that not all of these records will be included in the context window. The "in_context" column in these
# dataframes indicates whether the record is included in the context window.
# max_tokens: maximum number of tokens to use for the context window.


local_context_params = {
    "text_unit_prop": 0.5,
    "community_prop": 0.1,
    "conversation_history_max_turns": 5,
    "conversation_history_user_turns_only": True,
    "top_k_mapped_entities": 10,
    "top_k_relationships": 10,
    "include_entity_rank": True,
    "include_relationship_weight": True,
    "include_community_rank": False,
    "return_candidate_context": False,
    "embedding_vectorstore_key": EntityVectorStoreKey.ID,  # set this to EntityVectorStoreKey.TITLE if the vectorstore uses entity title as ids
    "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)
}

llm_params = {
    "max_tokens": 2_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 1000=1500)
    "temperature": 0.0,
}

# text_unit_prop: proportion of context window dedicated to related text units # community_prop: proportion of context window dedicated to community reports. # The remaining proportion is dedicated to entities and relationships. Sum of text_unit_prop and community_prop should be <= 1 # conversation_history_max_turns: maximum number of turns to include in the conversation history. # conversation_history_user_turns_only: if True, only include user queries in the conversation history. # top_k_mapped_entities: number of related entities to retrieve from the entity description embedding store. # top_k_relationships: control the number of out-of-network relationships to pull into the context window. # include_entity_rank: if True, include the entity rank in the entity table in the context window. Default entity rank = node degree. # include_relationship_weight: if True, include the relationship weight in the context window. # include_community_rank: if True, include the community rank in the context window. # return_candidate_context: if True, return a set of dataframes containing all candidate entity/relationship/covariate records that # could be relevant. Note that not all of these records will be included in the context window. The "in_context" column in these # dataframes indicates whether the record is included in the context window. # max_tokens: maximum number of tokens to use for the context window. local_context_params = { "text_unit_prop": 0.5, "community_prop": 0.1, "conversation_history_max_turns": 5, "conversation_history_user_turns_only": True, "top_k_mapped_entities": 10, "top_k_relationships": 10, "include_entity_rank": True, "include_relationship_weight": True, "include_community_rank": False, "return_candidate_context": False, "embedding_vectorstore_key": EntityVectorStoreKey.ID, # set this to EntityVectorStoreKey.TITLE if the vectorstore uses entity title as ids "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) } llm_params = { "max_tokens": 2_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 1000=1500) "temperature": 0.0, }

In [12]:

search_engine = LocalSearch(
    llm=llm,
    context_builder=context_builder,
    token_encoder=token_encoder,
    llm_params=llm_params,
    context_builder_params=local_context_params,
    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 = LocalSearch( llm=llm, context_builder=context_builder, token_encoder=token_encoder, llm_params=llm_params, context_builder_params=local_context_params, 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[12], line 3
      1 search_engine = LocalSearch(
      2     llm=llm,
----> 3     context_builder=context_builder,
      4     token_encoder=token_encoder,
      5     llm_params=llm_params,
      6     context_builder_params=local_context_params,
      7     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
      8 )

NameError: name 'context_builder' is not defined

Run local search on sample queries¶

In [13]:

result = await search_engine.asearch("Tell me about Agent Mercer")
print(result.response)

result = await search_engine.asearch("Tell me about Agent Mercer") print(result.response)

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[13], line 1
----> 1 result = await search_engine.asearch("Tell me about Agent Mercer")
      2 print(result.response)

NameError: name 'search_engine' is not defined

In [14]:

question = "Tell me about Dr. Jordan Hayes"
result = await search_engine.asearch(question)
print(result.response)

question = "Tell me about Dr. Jordan Hayes" result = await search_engine.asearch(question) print(result.response)

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[14], line 2
      1 question = "Tell me about Dr. Jordan Hayes"
----> 2 result = await search_engine.asearch(question)
      3 print(result.response)

NameError: name 'search_engine' is not defined

Inspecting the context data used to generate the response¶

In [15]:

result.context_data["entities"].head()

result.context_data["entities"].head()

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[15], line 1
----> 1 result.context_data["entities"].head()

NameError: name 'result' is not defined

In [16]:

result.context_data["relationships"].head()

result.context_data["relationships"].head()

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[16], line 1
----> 1 result.context_data["relationships"].head()

NameError: name 'result' is not defined

In [17]:

if "reports" in result.context_data:
    result.context_data["reports"].head()

if "reports" in result.context_data: result.context_data["reports"].head()

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[17], line 1
----> 1 if "reports" in result.context_data:
      2     result.context_data["reports"].head()

NameError: name 'result' is not defined

In [18]:

result.context_data["sources"].head()

result.context_data["sources"].head()

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[18], line 1
----> 1 result.context_data["sources"].head()

NameError: name 'result' is not defined

In [19]:

if "claims" in result.context_data:
    print(result.context_data["claims"].head())

if "claims" in result.context_data: print(result.context_data["claims"].head())

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[19], line 1
----> 1 if "claims" in result.context_data:
      2     print(result.context_data["claims"].head())

NameError: name 'result' is not defined

Question Generation¶

This function takes a list of user queries and generates the next candidate questions.

In [20]:

question_generator = LocalQuestionGen(
    llm=llm,
    context_builder=context_builder,
    token_encoder=token_encoder,
    llm_params=llm_params,
    context_builder_params=local_context_params,
)

question_generator = LocalQuestionGen( llm=llm, context_builder=context_builder, token_encoder=token_encoder, llm_params=llm_params, context_builder_params=local_context_params, )

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[20], line 3
      1 question_generator = LocalQuestionGen(
      2     llm=llm,
----> 3     context_builder=context_builder,
      4     token_encoder=token_encoder,
      5     llm_params=llm_params,
      6     context_builder_params=local_context_params,
      7 )

NameError: name 'context_builder' is not defined

In [21]:

question_history = [
    "Tell me about Agent Mercer",
    "What happens in Dulce military base?",
]
candidate_questions = await question_generator.agenerate(
    question_history=question_history, context_data=None, question_count=5
)
print(candidate_questions.response)

question_history = [ "Tell me about Agent Mercer", "What happens in Dulce military base?", ] candidate_questions = await question_generator.agenerate( question_history=question_history, context_data=None, question_count=5 ) print(candidate_questions.response)

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[21], line 5
      1 question_history = [
      2     "Tell me about Agent Mercer",
      3     "What happens in Dulce military base?",
      4 ]
----> 5 candidate_questions = await question_generator.agenerate(
      6     question_history=question_history, context_data=None, question_count=5
      7 )
      8 print(candidate_questions.response)

NameError: name 'question_generator' is not defined