wehub-resource-sync
6b7e6b44f1
gh-pages / build (push) Waiting to run
Python Publish (pypi) / Upload release to PyPI (push) Waiting to run
Spellcheck / spellcheck (push) Waiting to run
Python Build and Type Check / python-ci (ubuntu-latest, 3.11) (push) Has been cancelled
Python Build and Type Check / python-ci (ubuntu-latest, 3.13) (push) Has been cancelled
Python Build and Type Check / python-ci (windows-latest, 3.11) (push) Has been cancelled
Python Build and Type Check / python-ci (windows-latest, 3.13) (push) Has been cancelled
Python Integration Tests / python-ci (ubuntu-latest, 3.13) (push) Has been cancelled
Python Integration Tests / python-ci (windows-latest, 3.13) (push) Has been cancelled
Python Notebook Tests / python-ci (ubuntu-latest, 3.13) (push) Has been cancelled
Python Notebook Tests / python-ci (windows-latest, 3.13) (push) Has been cancelled
Python Smoke Tests / python-ci (ubuntu-latest, 3.13) (push) Has been cancelled
Python Smoke Tests / python-ci (windows-latest, 3.13) (push) Has been cancelled
Python Unit Tests / python-ci (ubuntu-latest, 3.13) (push) Has been cancelled
Python Unit Tests / python-ci (windows-latest, 3.13) (push) Has been cancelled
74 lines
3.1 KiB
Python
74 lines
3.1 KiB
Python
# Copyright (c) 2024 Microsoft Corporation.
|
|
# Licensed under the MIT License
|
|
|
|
from graphrag.data_model.schemas import COVARIATES_FINAL_COLUMNS
|
|
from graphrag.index.workflows.extract_covariates import (
|
|
run_workflow,
|
|
)
|
|
from graphrag_llm.config import LLMProviderType
|
|
from pandas.testing import assert_series_equal
|
|
|
|
from tests.unit.config.utils import get_default_graphrag_config
|
|
|
|
from .util import (
|
|
create_test_context,
|
|
load_test_table,
|
|
)
|
|
|
|
MOCK_LLM_RESPONSES = [
|
|
"""
|
|
(COMPANY A<|>GOVERNMENT AGENCY B<|>ANTI-COMPETITIVE PRACTICES<|>TRUE<|>2022-01-10T00:00:00<|>2022-01-10T00:00:00<|>Company A was found to engage in anti-competitive practices because it was fined for bid rigging in multiple public tenders published by Government Agency B according to an article published on 2022/01/10<|>According to an article published on 2022/01/10, Company A was fined for bid rigging while participating in multiple public tenders published by Government Agency B.)
|
|
""".strip()
|
|
]
|
|
|
|
|
|
async def test_extract_covariates():
|
|
input = load_test_table("text_units")
|
|
|
|
context = await create_test_context(
|
|
storage=["text_units"],
|
|
)
|
|
|
|
config = get_default_graphrag_config()
|
|
config.extract_claims.enabled = True
|
|
config.extract_claims.description = "description"
|
|
llm_settings = config.get_completion_model_config(
|
|
config.extract_claims.completion_model_id
|
|
)
|
|
llm_settings.type = LLMProviderType.MockLLM
|
|
llm_settings.mock_responses = MOCK_LLM_RESPONSES # type: ignore
|
|
|
|
await run_workflow(config, context)
|
|
|
|
actual = await context.output_table_provider.read_dataframe("covariates")
|
|
|
|
for column in COVARIATES_FINAL_COLUMNS:
|
|
assert column in actual.columns
|
|
|
|
# our mock only returns one covariate per text unit, so that's a 1:1 mapping versus the LLM-extracted content in the test data
|
|
assert len(actual) == len(input)
|
|
|
|
# assert all of the columns that covariates copied from the input
|
|
assert_series_equal(actual["text_unit_id"], input["id"], check_names=False)
|
|
|
|
# make sure the human ids are incrementing
|
|
assert actual["human_readable_id"][0] == 0
|
|
assert actual["human_readable_id"][1] == 1
|
|
|
|
# check that the mock data is parsed and inserted into the correct columns
|
|
assert actual["covariate_type"][0] == "claim"
|
|
assert actual["subject_id"][0] == "COMPANY A"
|
|
assert actual["object_id"][0] == "GOVERNMENT AGENCY B"
|
|
assert actual["type"][0] == "ANTI-COMPETITIVE PRACTICES"
|
|
assert actual["status"][0] == "TRUE"
|
|
assert actual["start_date"][0] == "2022-01-10T00:00:00"
|
|
assert actual["end_date"][0] == "2022-01-10T00:00:00"
|
|
assert (
|
|
actual["description"][0]
|
|
== "Company A was found to engage in anti-competitive practices because it was fined for bid rigging in multiple public tenders published by Government Agency B according to an article published on 2022/01/10"
|
|
)
|
|
assert (
|
|
actual["source_text"][0]
|
|
== "According to an article published on 2022/01/10, Company A was fined for bid rigging while participating in multiple public tenders published by Government Agency B."
|
|
)
|