Top-K and Quantile Aggregator over a Large File
Top-K and Quantile Aggregator over a Large File: a task in termgrade-bandcentral-gemma4-31b-bash: TermGrade Band-Central: the 151-task selection and its splits (Harbor dataset). Section 1: Top-K - One line per record: <rank ,<id ,<value ,<label - rank starts at 1 for the largest value - Records…
The task
**Section 1: Top-K** - One line per record: `<rank>,<id>,<value>,<label>` - `rank` starts at 1 for the largest value - Records sorted by `value` descending - Ties (same `value`) broken by `id` ascending - Exactly K records (if fewer than K total valid records exist, output all of them) - If no valid records exist,…