conda create -n graphrag3.10 python=3.10
oonda activate graphrag3.10

ollama run qwen2:7b

git clone git@github.com:microsoft/graphrag.git
cd graphrag
pip install -e.

mkdir -p ./graphrag_sophia/input  # 注意一定要建立一个input子文件夹，在创建索引的时候会去识别这个名字。如果是其他名字，可以在settings.yaml更改

python -m graphrag init --root ./graphrag_sophia

curl https://www.gutenberg.org/cache/epub/24022/pg24022.txt -o ./data/test/book.txt

llm:
model: qwen2:7b
api_base: http://127.0.0.1:11434/v1
model_supports_json: false
max_tokens: 32768

### This config file contains required core defaults that must be set, along with a handful of common optional settings.
### For a full list of available settings, see https://microsoft.github.io/graphrag/config/yaml/

### LLM settings ###
## There are a number of settings to tune the threading and token limits for LLM calls - check the docs.

models:
  default_chat_model:
    type: openai_chat # or azure_openai_chat
    api_base: http://127.0.0.1:11434/v1
    auth_type: api_key # or azure_managed_identity
    api_key: ollama # set this in the generated .env file
    model: qwen2:7b
    encoding_model: cl100k_base # automatically set by tiktoken if left undefined
    model_supports_json: true # recommended if this is available for your model.
    concurrent_requests: 25 # max number of simultaneous LLM requests allowed
    async_mode: threaded # or asyncio
    retry_strategy: native
    max_retries: 10
    tokens_per_minute: auto              # set to null to disable rate limiting
    requests_per_minute: auto            # set to null to disable rate limiting
  default_embedding_model:
    type: openai_embedding # or azure_openai_embedding
    api_base: http://127.0.0.1:8005/api/v1 
    auth_type: api_key # or azure_managed_identity
    api_key: ollama
    model: bge
    encoding_model: cl100k_base # automatically set by tiktoken if left undefined
    model_supports_json: true # recommended if this is available for your model.
    concurrent_requests: 25 # max number of simultaneous LLM requests allowed
    async_mode: threaded # or asyncio
    retry_strategy: native
    max_retries: 10
    tokens_per_minute: auto              # set to null to disable rate limiting
    requests_per_minute: auto            # set to null to disable rate limiting

### Input settings ###

input:
  type: file # or blob
  file_type: text # [csv, text, json]
  base_dir: "input"

chunks:
  size: 200 #1200
  overlap: 50 # 100 
  group_by_columns: [id]

### Output/storage settings ###
## If blob storage is specified in the following four sections,
## connection_string and container_name must be provided

output:
  type: file # [file, blob, cosmosdb]
  base_dir: "output"
    
cache:
  type: file # [file, blob, cosmosdb]
  base_dir: "cache"

reporting:
  type: file # [file, blob, cosmosdb]
  base_dir: "logs"

vector_store:
  default_vector_store:
    type: lancedb
    db_uri: output\lancedb
    container_name: default
    overwrite: True

### Workflow settings ###

embed_text:
  model_id: default_embedding_model
  vector_store_id: default_vector_store

extract_graph:
  model_id: default_chat_model
  prompt: "prompts/extract_graph.txt"
  entity_types: [organization,person,geo,event]
  max_gleanings: 1

summarize_descriptions:
  model_id: default_chat_model
  prompt: "prompts/summarize_descriptions.txt"
  max_length: 500

extract_graph_nlp:
  text_analyzer:
    extractor_type: regex_english # [regex_english, syntactic_parser, cfg]

cluster_graph:
  max_cluster_size: 10

extract_claims:
  enabled: false
  model_id: default_chat_model
  prompt: "prompts/extract_claims.txt"
  description: "Any claims or facts that could be relevant to information discovery."
  max_gleanings: 1

community_reports:
  model_id: default_chat_model
  graph_prompt: "prompts/community_report_graph.txt"
  text_prompt: "prompts/community_report_text.txt"
  max_length: 2000
  max_input_length: 8000

embed_graph:
  enabled: false # if true, will generate node2vec embeddings for nodes

umap:
  enabled: false # if true, will generate UMAP embeddings for nodes (embed_graph must also be enabled)

snapshots:
  graphml: false
  embeddings: false

### Query settings ###
## The prompt locations are required here, but each search method has a number of optional knobs that can be tuned.
## See the config docs: https://microsoft.github.io/graphrag/config/yaml/#query

local_search:
  chat_model_id: default_chat_model
  embedding_model_id: default_embedding_model
  prompt: "prompts/local_search_system_prompt.txt"

global_search:
  chat_model_id: default_chat_model
  map_prompt: "prompts/global_search_map_system_prompt.txt"
  reduce_prompt: "prompts/global_search_reduce_system_prompt.txt"
  knowledge_prompt: "prompts/global_search_knowledge_system_prompt.txt"

drift_search:
  chat_model_id: default_chat_model
  embedding_model_id: default_embedding_model
  prompt: "prompts/drift_search_system_prompt.txt"
  reduce_prompt: "prompts/drift_search_reduce_prompt.txt"

basic_search:
  chat_model_id: default_chat_model
  embedding_model_id: default_embedding_model
  prompt: "prompts/basic_search_system_prompt.txt"

graphrag query --root .\graphrag_sophia --method global --query "Please introduce the Forbidden City"

Warning: All map responses have score 0 (i.e., no relevant information found from the dataset), returning a canned 'I do not know' answer. You can try enabling `allow_general_knowledge` to encourage the LLM to incorporate relevant general knowledge, at the risk of increasing hallucinations.

SUCCESS: Global Search Response:
I am sorry but I am unable to answer this question given the provided data.

INFO: Vector Store Args: {
"default_vector_store": {
"type": "lancedb",
"db_uri": "C:\\Users\\Sophia\\graphrag\\graphrag_sophia\\output\\lancedb",
"url": null,
"audience": null,
"container_name": "==== REDACTED ====",
"database_name": null,
"overwrite": true
}
}

SUCCESS: Local Search Response:
The Forbidden City, officially known as the Palace Museum, is a historic palace complex located in the heart of Beijing, China. It was constructed during the Ming dynasty (1406-1420) and served as the imperial palace for 24 emperors over the course of the Ming and Qing dynasties (1368-1912). The Forbidden City covers an area of approximately 720,000 square meters and is surrounded by a large moat.

The complex is divided into two main sections: the Outer Court and the Inner Court. The Outer Court was where the emperor conducted official business and received foreign dignitants, while the Inner Court was reserved for the imperial family's private life, including the emperor, empresses, concubines, and their children.

The Forbidden City is renowned for its grand architecture, which showcases traditional Chinese building styles with intricate details such as roof tiles decorated with dragons, phoenixes, and other mythical creatures. The palace buildings are arranged in a symmetrical layout, reflecting the imperial ideology of harmony and balance.

Today, the Forbidden City serves as a museum that houses over 1 million artifacts and artworks from various dynasties. It is one of the most visited tourist attractions in China and is recognized as a UNESCO World Heritage Site for its cultural significance and historical value. The palace complex not only preserves the legacy of ancient Chinese architecture but also provides insights into the imperial lifestyle, court rituals, and the political structure of the Ming and Qing dynasties.

Visitors can explore various halls, palaces, gardens, and courtyards within the Forbidden City, each with its own unique history and stories. Some notable structures include the Hall of Supreme Harmony (Zhonghe Dian), the Hall of Central Harmony (Zhongheng Dian), the Hall of Earthly Tranquility (Taixing Dian), and the Palace of Heavenly Purity (Tianqing Gong). The palace also hosts special exhibitions that showcase Chinese art, calligraphy, ceramics, and other cultural treasures.