mirror of
https://github.com/microsoft/graphrag.git
synced 2026-01-14 09:07:20 +08:00
69ad36e735
Some checks failed
gh-pages / build (push) Has been cancelled
Python CI / python-ci (ubuntu-latest, 3.10) (push) Has been cancelled
Python CI / python-ci (ubuntu-latest, 3.11) (push) Has been cancelled
Python CI / python-ci (windows-latest, 3.10) (push) Has been cancelled
Python CI / python-ci (windows-latest, 3.11) (push) Has been cancelled
Python Integration Tests / python-ci (ubuntu-latest, 3.10) (push) Has been cancelled
Python Integration Tests / python-ci (windows-latest, 3.10) (push) Has been cancelled
Python Notebook Tests / python-ci (ubuntu-latest, 3.10) (push) Has been cancelled
Python Notebook Tests / python-ci (windows-latest, 3.10) (push) Has been cancelled
Python Publish (pypi) / Upload release to PyPI (push) Has been cancelled
Python Smoke Tests / python-ci (ubuntu-latest, 3.10) (push) Has been cancelled
Python Smoke Tests / python-ci (windows-latest, 3.10) (push) Has been cancelled
Spellcheck / spellcheck (push) Has been cancelled
* Fix all human_readable_id columns to start at 0 * Semver
81 lines
2.6 KiB
Python
81 lines
2.6 KiB
Python
# Copyright (c) 2024 Microsoft Corporation.
|
|
# Licensed under the MIT License
|
|
|
|
from graphrag.config.create_graphrag_config import create_graphrag_config
|
|
from graphrag.data_model.schemas import (
|
|
ENTITIES_FINAL_COLUMNS,
|
|
RELATIONSHIPS_FINAL_COLUMNS,
|
|
)
|
|
from graphrag.index.workflows.finalize_graph import (
|
|
run_workflow,
|
|
)
|
|
from graphrag.utils.storage import load_table_from_storage, write_table_to_storage
|
|
|
|
from .util import (
|
|
DEFAULT_MODEL_CONFIG,
|
|
create_test_context,
|
|
load_test_table,
|
|
)
|
|
|
|
|
|
async def test_finalize_graph():
|
|
context = await _prep_tables()
|
|
|
|
config = create_graphrag_config({"models": DEFAULT_MODEL_CONFIG})
|
|
|
|
await run_workflow(config, context)
|
|
|
|
nodes_actual = await load_table_from_storage("entities", context.output_storage)
|
|
edges_actual = await load_table_from_storage(
|
|
"relationships", context.output_storage
|
|
)
|
|
|
|
# x and y will be zero with the default configuration, because we do not embed/umap
|
|
assert nodes_actual["x"].sum() == 0
|
|
assert nodes_actual["y"].sum() == 0
|
|
|
|
for column in ENTITIES_FINAL_COLUMNS:
|
|
assert column in nodes_actual.columns
|
|
for column in RELATIONSHIPS_FINAL_COLUMNS:
|
|
assert column in edges_actual.columns
|
|
|
|
|
|
async def test_finalize_graph_umap():
|
|
context = await _prep_tables()
|
|
|
|
config = create_graphrag_config({"models": DEFAULT_MODEL_CONFIG})
|
|
|
|
config.embed_graph.enabled = True
|
|
config.umap.enabled = True
|
|
|
|
await run_workflow(config, context)
|
|
|
|
nodes_actual = await load_table_from_storage("entities", context.output_storage)
|
|
edges_actual = await load_table_from_storage(
|
|
"relationships", context.output_storage
|
|
)
|
|
|
|
# x and y should have some value other than zero due to umap
|
|
assert nodes_actual["x"].sum() != 0
|
|
assert nodes_actual["y"].sum() != 0
|
|
|
|
for column in ENTITIES_FINAL_COLUMNS:
|
|
assert column in nodes_actual.columns
|
|
for column in RELATIONSHIPS_FINAL_COLUMNS:
|
|
assert column in edges_actual.columns
|
|
|
|
|
|
async def _prep_tables():
|
|
context = await create_test_context(
|
|
storage=["entities", "relationships"],
|
|
)
|
|
|
|
# edit the tables to eliminate final fields that wouldn't be on the inputs
|
|
entities = load_test_table("entities")
|
|
entities.drop(columns=["x", "y", "degree"], inplace=True)
|
|
await write_table_to_storage(entities, "entities", context.output_storage)
|
|
relationships = load_test_table("relationships")
|
|
relationships.drop(columns=["combined_degree"], inplace=True)
|
|
await write_table_to_storage(relationships, "relationships", context.output_storage)
|
|
return context
|