minafarid
commented
5 years ago Hello and thank you for your interest,
HoloClean accepts CSV files as input. But we rely on a DBMS (postgres in this case) to store and query the data and perform various analytics that cannot be done on plain CSV or parquet files, which only store data. I would suggest using a docker for postgres if you are not familiar with its setup. Hopefully, this makes it easier for you to try HoloClean out.
Hi,
Would it be possible for someone to write a simple example that uses HoloClean with a basic csv, parquet file, or sqlite database instead of postgres? A HoloClean example without the postgres dependency, and even better than initially just uses pytorch-cpu, would make HoloClean feel accessible to a much larger user group.