mirror of
https://github.com/microsoft/graphrag.git
synced 2025-06-26 23:19:58 +00:00
1df89727c3
* Move covariate run conditional * All pipeline registration * Fix method name construction * Rename context storage -> output_storage * Rename OutputConfig as generic StorageConfig * Reuse Storage model under InputConfig * Move input storage creation out of document loading * Move document loading into workflows * Semver * Fix smoke test config for new workflows * Fix unit tests --------- Co-authored-by: Alonso Guevara <alonsog@microsoft.com>
69 lines
2.2 KiB
Python
69 lines
2.2 KiB
Python
# Copyright (c) 2024 Microsoft Corporation.
|
|
# Licensed under the MIT License
|
|
|
|
from graphrag.config.create_graphrag_config import create_graphrag_config
|
|
from graphrag.config.embeddings import (
|
|
all_embeddings,
|
|
)
|
|
from graphrag.config.enums import ModelType
|
|
from graphrag.index.operations.embed_text.embed_text import TextEmbedStrategyType
|
|
from graphrag.index.workflows.generate_text_embeddings import (
|
|
run_workflow,
|
|
)
|
|
from graphrag.utils.storage import load_table_from_storage
|
|
|
|
from .util import (
|
|
DEFAULT_MODEL_CONFIG,
|
|
create_test_context,
|
|
)
|
|
|
|
|
|
async def test_generate_text_embeddings():
|
|
context = await create_test_context(
|
|
storage=[
|
|
"documents",
|
|
"relationships",
|
|
"text_units",
|
|
"entities",
|
|
"community_reports",
|
|
]
|
|
)
|
|
|
|
config = create_graphrag_config({"models": DEFAULT_MODEL_CONFIG})
|
|
llm_settings = config.get_language_model_config(
|
|
config.embed_text.model_id
|
|
).model_dump()
|
|
llm_settings["type"] = ModelType.MockEmbedding
|
|
|
|
config.embed_text.strategy = {
|
|
"type": TextEmbedStrategyType.openai,
|
|
"llm": llm_settings,
|
|
}
|
|
config.embed_text.names = list(all_embeddings)
|
|
config.snapshots.embeddings = True
|
|
|
|
await run_workflow(config, context)
|
|
|
|
parquet_files = context.output_storage.keys()
|
|
|
|
for field in all_embeddings:
|
|
assert f"embeddings.{field}.parquet" in parquet_files
|
|
|
|
# entity description should always be here, let's assert its format
|
|
entity_description_embeddings = await load_table_from_storage(
|
|
"embeddings.entity.description", context.output_storage
|
|
)
|
|
|
|
assert len(entity_description_embeddings.columns) == 2
|
|
assert "id" in entity_description_embeddings.columns
|
|
assert "embedding" in entity_description_embeddings.columns
|
|
|
|
# every other embedding is optional but we've turned them all on, so check a random one
|
|
document_text_embeddings = await load_table_from_storage(
|
|
"embeddings.document.text", context.output_storage
|
|
)
|
|
|
|
assert len(document_text_embeddings.columns) == 2
|
|
assert "id" in document_text_embeddings.columns
|
|
assert "embedding" in document_text_embeddings.columns
|