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.input.loaders.dfs import (
    store_entity_semantic_embeddings,
)
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.input.loaders.dfs import ( store_entity_semantic_embeddings, ) 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="entity_description_embeddings",
)
description_embedding_store.connect(db_uri=LANCEDB_URI)
entity_description_embeddings = store_entity_semantic_embeddings(
    entities=entities, vectorstore=description_embedding_store
)

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="entity_description_embeddings", ) description_embedding_store.connect(db_uri=LANCEDB_URI) entity_description_embeddings = store_entity_semantic_embeddings( entities=entities, vectorstore=description_embedding_store ) print(f"Entity count: {len(entity_df)}") entity_df.head()

Entity count: 434

[2024-10-24T17:21:23Z WARN  lance::dataset] No existing dataset at /home/runner/work/graphrag/graphrag/docs/examples_notebooks/inputs/operation dulce/lancedb/entity_description_embeddings.lance, it will be created

Out[4]:

	level	title	type	description	source_id	community	degree	human_readable_id	id	size	graph_embedding	entity_type	top_level_node_id	x	y
0	0	ALEX MERCER	PERSON	Alex Mercer is a character with a military bac...	00fafabae48948779fee2afe600f5143,1e433d6b30887...	1	9	0	b45241d70f0e43fca764df95b2b81f77	9	None	None	b45241d70f0e43fca764df95b2b81f77	0	0
1	0	TAYLOR CRUZ	PERSON	Taylor Cruz is a character who plays a pivotal...	00fafabae48948779fee2afe600f5143,1e433d6b30887...	1	12	1	4119fd06010c494caa07f439b333f4c5	12	None	None	4119fd06010c494caa07f439b333f4c5	0	0
2	0	JORDAN HAYES	PERSON	Dr. Jordan Hayes is a key character in a narra...	00fafabae48948779fee2afe600f5143,2cf7a230c367a...	1	9	2	d3835bf3dda84ead99deadbeac5d0d7d	9	None	None	d3835bf3dda84ead99deadbeac5d0d7d	0	0
3	0	SAM RIVERA	PERSON	Sam Rivera is a character renowned for their t...	00fafabae48948779fee2afe600f5143,1e433d6b30887...	1	15	3	077d2820ae1845bcbb1803379a3d1eae	15	None	None	077d2820ae1845bcbb1803379a3d1eae	0	0
4	0	PARANORMAL MILITARY SQUAD	ORGANIZATION	The Paranormal Military Squad is an elite clan...	00fafabae48948779fee2afe600f5143,1e433d6b30887...	1	14	4	3671ea0dd4e84c1a9b02c5ab2c8f4bac	14	None	None	3671ea0dd4e84c1a9b02c5ab2c8f4bac	0	0

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()

Relationship count: 276

Out[5]:

	source	target	weight	description	text_unit_ids	id	human_readable_id	source_degree	target_degree	rank
0	ALEX MERCER	TAYLOR CRUZ	7.0	Alex Mercer and Taylor Cruz are integral membe...	[00fafabae48948779fee2afe600f5143, 1e433d6b308...	b35c3d1a7daa4924b6bdb58bc69c354d	0	9	12	21
1	ALEX MERCER	TAYLOR CRUZ	7.0	Alex Mercer and Taylor Cruz are integral membe...	[00fafabae48948779fee2afe600f5143, 1e433d6b308...	b35c3d1a7daa4924b6bdb58bc69c354d	0	9	12	21
2	ALEX MERCER	TAYLOR CRUZ	7.0	Alex Mercer and Taylor Cruz are integral membe...	[00fafabae48948779fee2afe600f5143, 1e433d6b308...	b35c3d1a7daa4924b6bdb58bc69c354d	0	9	12	21
3	ALEX MERCER	TAYLOR CRUZ	7.0	Alex Mercer and Taylor Cruz are integral membe...	[00fafabae48948779fee2afe600f5143, 1e433d6b308...	b35c3d1a7daa4924b6bdb58bc69c354d	0	9	12	21
4	ALEX MERCER	JORDAN HAYES	6.0	Alex Mercer and Jordan Hayes are colleagues wh...	[00fafabae48948779fee2afe600f5143, 2cf7a230c36...	a97e2ecd870944cfbe71c79bc0fcc752	1	9	9	18

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: 89

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: 6

Out[7]:

	community	full_content	level	rank	title	rank_explanation	summary	findings	full_content_json	id
0	4	# Dulce Base and the Paranormal Military Squad...	1	8.5	Dulce Base and the Paranormal Military Squad: ...	The impact severity rating is high due to the ...	The community is centered around Dulce Base, a...	[{'explanation': 'Dulce Base is the primary lo...	{\n "title": "Dulce Base and the Paranormal...	6f8ba6b6-506e-46c1-83ce-982d59622554
1	5	# Sam Rivera and the Paranormal Military Squad...	1	7.5	Sam Rivera and the Paranormal Military Squad a...	The impact severity rating is high due to the ...	The community is centered around Sam Rivera, a...	[{'explanation': 'Sam Rivera is recognized for...	{\n "title": "Sam Rivera and the Paranormal...	418f4536-d673-4212-8a7c-ca1aac547d0f
2	0	# Dulce Base and the Paranormal Military Squad...	0	8.5	Dulce Base and the Paranormal Military Squad Team	The impact severity rating is high due to the ...	Dulce Base serves as the operational hub for t...	[{'explanation': 'The Paranormal Military Squa...	{\n "title": "Dulce Base and the Paranormal...	251df57e-fd49-49a7-b262-ccaff95d7a51
3	1	# Dulce Base and the Paranormal Military Squad...	0	8.5	Dulce Base and the Paranormal Military Squad: ...	The impact severity rating is high due to the ...	The community is centered around Dulce Base, a...	[{'explanation': 'Sam Rivera is recognized for...	{\n "title": "Dulce Base and the Paranormal...	6e536385-8056-4a82-8670-c0ccaf007fb4
4	2	# Dulce Base: Extraterrestrial Research and Co...	0	8.5	Dulce Base: Extraterrestrial Research and Comm...	The impact severity rating is high due to the ...	Dulce Base is a highly classified facility in ...	[{'explanation': 'Dulce Base is the epicenter ...	{\n "title": "Dulce Base: Extraterrestrial ...	660c8fba-b65f-4fcd-9c98-203600cf1981

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: 12

Out[8]:

	id	text	n_tokens	document_ids	entity_ids	relationship_ids	covariate_ids
0	2cf7a230c367a2dfaf0fc3c903eb8948	# Operation: Dulce\n\n## Chapter 1\n\nThe thru...	2500	[958fdd043f17ade63cb13570b59df295]	[b45241d70f0e43fca764df95b2b81f77, 4119fd06010...	[b35c3d1a7daa4924b6bdb58bc69c354d, a97e2ecd870...	[ad5a2020-cdec-4982-acdf-dbe5ee530066, 9d8a0fe...
1	6d1255303acb7c9dc951cb0f5fc3042c	be the same.\n\n\*\n\nThe sense of foreboding...	2500	[958fdd043f17ade63cb13570b59df295]	[b45241d70f0e43fca764df95b2b81f77, 4119fd06010...	[b35c3d1a7daa4924b6bdb58bc69c354d, a97e2ecd870...	[5d1c9126-c48d-4755-9f9c-f739c823f95f, ec64a42...
2	e841f178310356740b2ee9101d12c97f	. "Your take on these signal inconsistencies?"...	2500	[958fdd043f17ade63cb13570b59df295]	[b45241d70f0e43fca764df95b2b81f77, 4119fd06010...	[b35c3d1a7daa4924b6bdb58bc69c354d, a97e2ecd870...	[0b22a34b-32e9-46a4-a0e8-d3d5466eba15, 7e14972...
3	f36d96862b9366d7240b5c7ceb04f12b	, absorbed in the bewilderment of contact, whi...	2500	[958fdd043f17ade63cb13570b59df295]	[b45241d70f0e43fca764df95b2b81f77, 4119fd06010...	[b35c3d1a7daa4924b6bdb58bc69c354d, 09f18f81442...	[9cd6d645-ab97-4b39-b02e-647cea9b5545, 50dc124...
4	f7d43808d2fb452cd953bf50c6de6bd4	were at once coherent and enigmatic: "*Voyage...	2500	[958fdd043f17ade63cb13570b59df295]	[b45241d70f0e43fca764df95b2b81f77, 4119fd06010...	[e02be3e37ca0454883a4c1fd859c24bb, 1dbc51475cb...	[87cf5900-6211-4e04-9115-50f3617c88b4]

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, )

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 )

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)

### Overview of Agent Alex Mercer

Agent Alex Mercer is a central figure within the Paranormal Military Squad Team at Dulce Base, where he plays a pivotal role in the team's operations and mission objectives. His responsibilities are multifaceted, encompassing leadership, strategic oversight, and direct involvement in the analysis and interpretation of extraterrestrial signals. Mercer's military background and experience are crucial to his role, as he guides the team through complex scenarios involving potential first contact with alien intelligence [Data: Entities (0, 209); Relationships (5, 8, 65)].

### Leadership and Responsibilities

As a leader, Alex Mercer is instrumental in overseeing the team's efforts to engage with extraterrestrial intelligence. His role involves ensuring a cautious and strategic approach to interspecies communication, which is vital for the success of their mission. Mercer's leadership is characterized by a mix of concern and anticipation, reflecting the gravity of the mission at hand. He is responsible for guiding the team's response to extraterrestrial contact, ensuring that the engagement is handled with care and foresight [Data: Entities (0); Relationships (5, 8, 6)].

### Collaboration and Team Dynamics

Agent Mercer works closely with other key members of the Paranormal Military Squad, such as Dr. Jordan Hayes and Sam Rivera. His collaboration with Dr. Hayes is particularly significant, as they jointly focus on decrypting and communicating with extraterrestrial intelligence. This partnership is built on mutual respect and recognition of each other's analytical skills, which are essential for the team's success. Mercer's interactions with Sam Rivera highlight his role as a mentor, providing guidance and fostering a relationship based on intuition and trust [Data: Reports (0); Relationships (1, 4, 2)].

### Involvement in Extraterrestrial Communication

Mercer's involvement in the decryption and analysis of alien signals is a critical aspect of his role. He is actively engaged in interpreting these signals, contributing to the understanding of an extraterrestrial society. His efforts are not only focused on overseeing the team but also on unraveling galactic mysteries and engaging with alien signals. This involvement underscores his position as a key figure in the mission's hypothesis and decision-making processes [Data: Claims (73, 85, 82); Reports (0)].

In summary, Agent Alex Mercer is a vital member of the Paranormal Military Squad Team, whose leadership and expertise are crucial to the team's mission at Dulce Base. His role in guiding the team through the complexities of extraterrestrial communication and his collaboration with other team members highlight his importance in the broader objectives of their work.

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)

### Overview of Dr. Jordan Hayes

Dr. Jordan Hayes is a prominent member of the Paranormal Military Squad, a specialized team based at Dulce Base, dedicated to the study and communication with extraterrestrial entities. Dr. Hayes plays a crucial role in the team's mission, focusing on deciphering alien code and interpreting extraterrestrial patterns. This expertise is vital to the squad's efforts in understanding and interacting with alien intelligence [Data: Entities (104, 2); Reports (0)].

### Role and Responsibilities

Dr. Hayes is deeply involved in the analysis and interpretation of alien signals, which includes isolating signal harmonics, decrypting alien messages, and interpreting these signals for further analysis. This work is central to the team's operations at Dulce Base, where Dr. Hayes collaborates closely with other team members, such as Agent Alex Mercer, to manage interspecies communication [Data: Entities (2, 166, 192); Relationships (1, 4, 26, 67)].

### Expertise and Approach

Known for an analytical and skeptical mindset, Dr. Hayes emphasizes empirical evidence and adaptability in the face of the unknown. This approach is particularly evident in their work on decryption algorithms and signal analysis, which are crucial for deciphering extraterrestrial communications. Dr. Hayes' role is pivotal in the team's focus on analyzing and interpreting alien signals, which is a cornerstone of their operation at Dulce Base [Data: Entities (2, 180, 124); Claims (12, 54, 68)].

### Collaboration and Impact

Dr. Hayes works closely with colleagues such as Sam Rivera and Taylor Cruz, contributing to the team's collective effort in deciphering alien code and preparing for first contact. The collaboration with Agent Alex Mercer is particularly significant, as they share a mutual respect and understanding of each other's strengths and the mission's significance. This teamwork is essential for the success of their mission, as they bring their expertise to the forefront of paranormal military endeavors [Data: Reports (0); Relationships (21, 9, 1, 4, 67)].

### Significance of Work

The work of Dr. Hayes and the Paranormal Military Squad is not just about scientific discovery but also about preparing humanity for potential interstellar communication. The team's efforts in analyzing extraterrestrial patterns and first contact data are indicative of their role in predicting extraterrestrial intentions and actions. Dr. Hayes' contributions are crucial in this historic effort, which could mark the beginning of a new era in human-alien relations [Data: Reports (0); Claims (84, 54, 61)].

In summary, Dr. Jordan Hayes is a key figure in the Paranormal Military Squad, bringing a critical and methodical approach to the team's extraordinary endeavors in understanding and interacting with alien entities. Their work at Dulce Base is central to the mission of deciphering extraterrestrial signals and preparing for potential interspecies communication.

Inspecting the context data used to generate the response¶

In [15]:

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

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

Out[15]:

	id	entity	description	number of relationships	in_context
0	104	DR. JORDAN HAYES	Dr. Jordan Hayes is a member of the Paranormal...	7	True
1	153	EXPERTISE	The specialized knowledge or skill that team m...	0	True
2	2	JORDAN HAYES	Dr. Jordan Hayes is a key character in a narra...	9	True
3	148	DATA ANALYSIS	The process of examining and interpreting info...	0	True
4	140	STAR ALIGNMENTS	Celestial configurations that Dr. Jordan Hayes...	0	True

In [16]:

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

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

Out[16]:

	id	source	target	description	weight	rank	links	in_context
0	21	JORDAN HAYES	SAM RIVERA	Jordan Hayes and Sam Rivera are colleagues at ...	5.0	24	2	True
1	22	JORDAN HAYES	PARANORMAL MILITARY SQUAD	Jordan Hayes is a member of the Paranormal Mil...	5.0	23	2	True
2	34	SAM RIVERA	DR. JORDAN HAYES	Dr. Jordan Hayes and Sam Rivera are colleagues...	1.0	22	2	True
3	9	TAYLOR CRUZ	JORDAN HAYES	Taylor Cruz and Jordan Hayes are both integral...	6.0	21	2	True
4	40	PARANORMAL MILITARY SQUAD	DR. JORDAN HAYES	Dr. Jordan Hayes is a key member of the Parano...	2.0	21	2	True

In [17]:

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

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

Out[17]:

	id	title	content
0	0	Dulce Base and the Paranormal Military Squad Team	# Dulce Base and the Paranormal Military Squad...

In [18]:

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

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

Out[18]:

	id	text
0	5	the universe.\n\nIn a symphony of clicks and ...
1	11	, the sentinel within them ever alert.\n\nAlex...
2	3	, absorbed in the bewilderment of contact, whi...

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())

   id            entity object_id     status start_date end_date  \
0  12  DR. JORDAN HAYES      NONE       TRUE       NONE     NONE   
1  13  DR. JORDAN HAYES      NONE  SUSPECTED       NONE     NONE   
2  18  DR. JORDAN HAYES      ALEX       TRUE       NONE     NONE   
3  49  DR. JORDAN HAYES      NONE  SUSPECTED       NONE     NONE   
4  74  DR. JORDAN HAYES      NONE  SUSPECTED       NONE     NONE   

                                         description  in_context  
0  Dr. Jordan Hayes contemplates their skepticism...        True  
1  Dr. Jordan Hayes mused over the layers of data...        True  
2  Dr. Jordan Hayes and Alex discovered a panel h...        True  
3  Dr. Jordan Hayes was focused on deciphering al...        True  
4  Dr. Jordan Hayes is analyzing the evolving ali...        True  

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, )

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)

['- What is the role of Agent Alex Mercer in Operation: Dulce?', '- How does the Paranormal Military Squad interact with extraterrestrial intelligence at Dulce Base?', '- What are the main objectives of Operation: Dulce at the Dulce Military Base?', '- How does the environment of the Dulce Military Base affect the team members involved in the operation?', "- What is the significance of New Mexico's location for the Dulce Military Base and Operation: Dulce?"]