Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.cognee.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
cognee.improve()

async def improve(
    dataset: Union[str, UUID] = "main_dataset",
    *,
    run_in_background: bool = False,
    node_name: Optional[List[str]] = None,
    session_ids: Optional[List[str]] = None,
    **kwargs,
)

​
Description

improve() enriches an existing graph after ingestion.
  • Without session_ids, it runs the normal enrichment pass over the dataset.
  • With session_ids, it can also apply feedback weights, persist session Q&A into the graph, and sync enriched graph context back into sessions.
For the full behavior walkthrough, see Improve.

​
Parameters

​
dataset
Union[str, UUID]
default:"'main_dataset'"
Dataset name or UUID to improve.
​
run_in_background
bool
default:"False"
Starts the improvement pipeline asynchronously.
​
node_name
Optional[List[str]]
default:"None"
Restricts improvement to specific named entities or node sets.
​
session_ids
Optional[List[str]]
default:"None"
Session IDs whose feedback and Q&A content should be bridged into the permanent graph.

​
Additional keyword options

OptionTypeWhat it does
extraction_taskslistOverrides the extraction task set used during enrichment.
enrichment_taskslistOverrides the enrichment task set used during enrichment.
dataAnySupplies explicit data to advanced improvement pipelines when supported.
node_typeTypeChanges which node type the enrichment pass targets.
userobjectRuns the operation under a specific user context.
vector_db_configdictOverrides vector database configuration for this call.
graph_db_configdictOverrides graph database configuration for this call.
feedback_alphafloatControls how strongly session feedback changes graph weights.

​
Return value

improve() returns the pipeline result from the enrichment pass, the same underlying shape used by legacy memify().

​
Examples

import cognee

await cognee.improve(
    dataset="product_docs",
    session_ids=["support_chat_7"],
)
See also memify() if you need lower-level control over the legacy enrichment pipeline.