3 ms·
Show HN: Finding errors in datasets with Similarity Search
The process of splitting or labeling data is error-prone, and these errors can be very costly. George Panchuk, ML Engineer at Qdrant, shares how to define misplaced items with similarity search and avoid errors in categorized datasets.