microsoft
create_final_covariates.parquet not generated[Bug]: <title>
Describe the bug
create_final_covariates.parquet not generated
Steps to reproduce
No response
Expected Behavior
No response
GraphRAG Config Used
No response
Logs and screenshots
No response
Additional Information
- GraphRAG Version:
- Operating System:
- Python Version:
- Related Issues:
Hi! Please provide more information about the issue. Can you please add the config used?
if you're using the default indexing pipeline it does not generate a create_final_covariates.parquet anymore. Set the covariates to None or don't include it when defining a LocalSearchMixedContext class.
if you're using the default indexing pipeline it does not generate a create_final_covariates.parquet anymore. Set the covariates to [] or don't include it when defining a LocalSearchMixedContext class.
What happens if I don't include covariates? How do I generate covariates?"
if you're using the default indexing pipeline it does not generate a create_final_covariates.parquet anymore. Set the covariates to [] or don't include it when defining a LocalSearchMixedContext class.
What happens if I don't include covariates? How do I generate covariates?"
Covariates are claims associated with the extracted entities. I'm not entirely sure why they decided to disable the covariate file. I'm guessing it's because of reducing LLM calls? GraphRAG is very expensive to run even on small indexing tasks.
Anyways, if you want to generate covariates, you can set GRAPHRAG_CLAIM_EXTRACTION_ENABLED to True in the .env file in your data project root. It should then generate the create_final_covariates.parquet file.
Thanks for your help @zanderjiang Absolutely correct.
Also, @win4r you can turn it on here:
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: 1
Just uncomment the enabled line in your settings.yaml file. I'll resolve the issue, but please reopen if this doesn't work
After enable claim_extraction, I got covariate_type keyerror in the step "create final covariates"
启用claim_extraction后，我在“create final covariates”步骤中收到了covariate_type keyerror
how do you solve this problem
