2
I spent months training an AI model on the wrong data set
I was working on a project to predict local traffic patterns in Austin, and after three months, my model was only 52% accurate. I found out last week that I had been using general US traffic data instead of the specific Austin city data I had collected. A friend pointed out the mismatch when he saw my code. I had to start over with the right data, which set me back weeks. Has anyone else made a basic data mistake that cost them a lot of time?
3 comments
Log in to join the discussion
Log In3 Comments
finley_wells6618d ago
Omg that's brutal. Spent a whole weekend cleaning a dataset once only to realize I was filtering by the wrong date column the entire time. Felt like such a waste.
4
olivia_hernandez16d ago
My friend Jake did that with 10,000 customer records.
3