search

microsoft / graphrag

A modular graph-based Retrieval-Augmented Generation (RAG) system
https://microsoft.github.io/graphrag/
MIT License
13.83k stars 1.2k forks source link

Error in Leiden Algorithm in create_base_entity_graph #515

Open as1078 opened 3 weeks ago

as1078 commented 3 weeks ago

Describe the issue

I got an empty network when doing the Leiden clustering algorithm as follows: {"type": "error", "data": "Error executing verb \"cluster_graph\" in create_base_entity_graph: EmptyNetworkError", "stack": ... leiden.EmptyNetworkError: EmptyNetworkError\n", "source": "EmptyNetworkError", "details": null}

When opening my parquet files for each step in pandas, there is only an entity_graph column with an incomplete graphml URL. I saw on other posts that there should also be a clustered_graph column, but there is none for me. When I look in the cache directory however, both entity_extraction and summarize_descriptions have valid JSON results, so I'm not sure how exactly the graph became empty. My data is a set of .txt files of US Congressional hearings, and I previously used the prompt autotune feature to customize prompts to my data.

Steps to reproduce

joint-20240710T193325Z-001.zip To generate results, I simply ran the init command followed by !python -m graphrag.prompt_tune --root ./ragtest --domain "US congress hearings" and then !python -m graphrag.index --verbose --root ./ragtest

GraphRAG Config Used

encoding_model: cl100k_base skip_workflows: [] llm: api_key: ${GRAPHRAG_API_KEY} type: openai_chat # or azure_openai_chat model: gpt-4-turbo-preview model_supports_json: true # recommended if this is available for your model.

max_tokens: 4000

request_timeout: 180.0

api_base: https://.openai.azure.com

api_version: 2024-02-15-preview

organization:

deployment_name:

tokens_per_minute: 150_000 # set a leaky bucket throttle

requests_per_minute: 10_000 # set a leaky bucket throttle

max_retries: 10

max_retry_wait: 10.0

sleep_on_rate_limit_recommendation: true # whether to sleep when azure suggests wait-times

concurrent_requests: 25 # the number of parallel inflight requests that may be made

parallelization: stagger: 0.3

num_threads: 50 # the number of threads to use for parallel processing

async_mode: threaded # or asyncio

embeddings:

parallelization: override the global parallelization settings for embeddings

async_mode: threaded # or asyncio llm: api_key: ${GRAPHRAG_API_KEY} type: openai_embedding # or azure_openai_embedding model: text-embedding-3-small

api_base: https://.openai.azure.com

# api_version: 2024-02-15-preview
# organization: <organization_id>
# deployment_name: <azure_model_deployment_name>
# tokens_per_minute: 150_000 # set a leaky bucket throttle
# requests_per_minute: 10_000 # set a leaky bucket throttle
# max_retries: 10
# max_retry_wait: 10.0
# sleep_on_rate_limit_recommendation: true # whether to sleep when azure suggests wait-times
# concurrent_requests: 25 # the number of parallel inflight requests that may be made
# batch_size: 16 # the number of documents to send in a single request
# batch_max_tokens: 8191 # the maximum number of tokens to send in a single request
# target: required # or optional

chunks: size: 300 overlap: 100 group_by_columns: [id] # by default, we don't allow chunks to cross documents

input: type: file # or blob file_type: text # or csv base_dir: "input" file_encoding: utf-8 file_pattern: ".*\.txt$"

cache: type: file # or blob base_dir: "cache"

connection_string:

container_name:

storage: type: file # or blob base_dir: "output/${timestamp}/artifacts"

connection_string:

container_name:

reporting: type: file # or console, blob base_dir: "output/${timestamp}/reports"

connection_string:

container_name:

entity_extraction:

llm: override the global llm settings for this task

parallelization: override the global parallelization settings for this task

async_mode: override the global async_mode settings for this task

prompt: "prompts/entity_extraction.txt" entity_types: [organization,person,geo,event] max_gleanings: 0

summarize_descriptions:

llm: override the global llm settings for this task

parallelization: override the global parallelization settings for this task

async_mode: override the global async_mode settings for this task

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

claim_extraction:

llm: override the global llm settings for this task

parallelization: override the global parallelization settings for this task

async_mode: override the global async_mode settings for this task

enabled: true

prompt: "prompts/claim_extraction.txt" description: "Any claims or facts that could be relevant to information discovery." max_gleanings: 0

community_report:

llm: override the global llm settings for this task

parallelization: override the global parallelization settings for this task

async_mode: override the global async_mode settings for this task

prompt: "prompts/community_report.txt" max_length: 2000 max_input_length: 8000

cluster_graph: max_cluster_size: 10

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

num_walks: 10

walk_length: 40

window_size: 2

iterations: 3

random_seed: 597832

umap: enabled: false # if true, will generate UMAP embeddings for nodes

snapshots: graphml: false raw_entities: false top_level_nodes: false

local_search:

text_unit_prop: 0.5

community_prop: 0.1

conversation_history_max_turns: 5

top_k_mapped_entities: 10

top_k_relationships: 10

max_tokens: 12000

global_search:

max_tokens: 12000

data_max_tokens: 12000

map_max_tokens: 1000

reduce_max_tokens: 2000

concurrency: 32

Logs and screenshots

Logs.json {"type": "error", "data": "Error executing verb \"cluster_graph\" in create_base_entity_graph: EmptyNetworkError", "stack": "Traceback (most recent call last):\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/datashaper/workflow/workflow.py\", line 410, in _execute_verb\n result = node.verb.func(verb_args)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/cluster_graph.py\", line 61, in cluster_graph\n results = output_df[column].apply(lambda graph: run_layout(strategy, graph))\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/series.py\", line 4924, in apply\n ).apply()\n ^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/apply.py\", line 1427, in apply\n return self.apply_standard()\n ^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/apply.py\", line 1507, in apply_standard\n mapped = obj._map_values(\n ^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/base.py\", line 921, in _map_values\n return algorithms.map_array(arr, mapper, na_action=na_action, convert=convert)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/algorithms.py\", line 1743, in map_array\n return lib.map_infer(values, mapper, convert=convert)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"lib.pyx\", line 2972, in pandas._libs.lib.map_infer\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/cluster_graph.py\", line 61, in \n results = output_df[column].apply(lambda graph: run_layout(strategy, graph))\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/cluster_graph.py\", line 167, in run_layout\n clusters = run_leiden(graph, strategy)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/strategies/leiden.py\", line 26, in run\n node_id_to_community_map = _compute_leiden_communities(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/strategies/leiden.py\", line 61, in _compute_leiden_communities\n community_mapping = hierarchical_leiden(\n ^^^^^^^^^^^^^^^^^^^^\n File \"<@beartype(graspologic.partition.leiden.hierarchical_leiden) at 0x32b439d00>\", line 304, in hierarchical_leiden\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graspologic/partition/leiden.py\", line 588, in hierarchical_leiden\n hierarchical_clusters_native = gn.hierarchical_leiden(\n ^^^^^^^^^^^^^^^^^^^^^^^\nleiden.EmptyNetworkError: EmptyNetworkError\n", "source": "EmptyNetworkError", "details": null} {"type": "error", "data": "Error running pipeline!", "stack": "Traceback (most recent call last):\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/run.py\", line 323, in run_pipeline\n result = await workflow.run(context, callbacks)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/datashaper/workflow/workflow.py\", line 369, in run\n timing = await self._execute_verb(node, context, callbacks)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/datashaper/workflow/workflow.py\", line 410, in _execute_verb\n result = node.verb.func(verb_args)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/cluster_graph.py\", line 61, in cluster_graph\n results = output_df[column].apply(lambda graph: run_layout(strategy, graph))\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/series.py\", line 4924, in apply\n ).apply()\n ^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/apply.py\", line 1427, in apply\n return self.apply_standard()\n ^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/apply.py\", line 1507, in apply_standard\n mapped = obj._map_values(\n ^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/base.py\", line 921, in _map_values\n return algorithms.map_array(arr, mapper, na_action=na_action, convert=convert)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/pandas/core/algorithms.py\", line 1743, in map_array\n return lib.map_infer(values, mapper, convert=convert)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"lib.pyx\", line 2972, in pandas._libs.lib.map_infer\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/cluster_graph.py\", line 61, in \n results = output_df[column].apply(lambda graph: run_layout(strategy, graph))\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/cluster_graph.py\", line 167, in run_layout\n clusters = run_leiden(graph, strategy)\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/strategies/leiden.py\", line 26, in run\n node_id_to_community_map = _compute_leiden_communities(\n ^^^^^^^^^^^^^^^^^^^^^^^^^^^^\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graphrag/index/verbs/graph/clustering/strategies/leiden.py\", line 61, in _compute_leiden_communities\n community_mapping = hierarchical_leiden(\n ^^^^^^^^^^^^^^^^^^^^\n File \"<@beartype(graspologic.partition.leiden.hierarchical_leiden) at 0x32b439d00>\", line 304, in hierarchical_leiden\n File \"/Users/amansingh/anaconda3/lib/python3.11/site-packages/graspologic/partition/leiden.py\", line 588, in hierarchical_leiden\n hierarchical_clusters_native = gn.hierarchical_leiden(\n ^^^^^^^^^^^^^^^^^^^^^^^\nleiden.EmptyNetworkError: EmptyNetworkError\n", "source": "EmptyNetworkError", "details": null}

Additional Information

AlonsoGuevara commented 3 weeks ago

HI @as1078

It seems entity extraction process failed and yield an empty graph. Could you please share your log file?

as1078 commented 3 weeks ago

Sure, since my log file is too large to upload, I have uploaded a portion of it here It seems that all other errors besides the clustering one were rate limit errors, which I thought were dealt with by GraphRAG through waiting before submitting another API request. I excluded the clustering errors that were put in above. logs.json

fire commented 3 weeks ago

I have this error too. I noticed that my generated prompts were missing a ) in the entity extraction.

as1078 commented 3 weeks ago

Just noticed I had the same issue. Thanks!

fire commented 3 weeks ago

On less performant models like the phi-3 https://github.com/microsoft/graphrag/pull/503 was able to repair the json. I did not test with prompt rewrite.

jiangjingzhi2003 commented 2 weeks ago

I have this error too. I noticed that my generated prompts were missing a ) in the entity extraction.

I still have the same error after I fix my generated prompts for entity extraction, does anyone know what might be the cause?

github-actions[bot] commented 1 week ago

This issue has been marked stale due to inactivity after repo maintainer or community member responses that request more information or suggest a solution. It will be closed after five additional days.