luispoveda93's picture
download
raw
1 kB
import json, collections
allr = [json.loads(l) for l in open('/work/questions_all.jsonl')]
pilot = [json.loads(l) for l in open('/work/pilot_questions.jsonl')]
print('=== COUNTS SUMMARY ===')
print('questions_all.jsonl records:', len(allr))
print('exams:', dict(collections.Counter(r['exam'] for r in allr)))
print('has_exhibit:', sum(r['has_exhibit'] for r in allr))
print('DRAG DROP / empty correct:', sum(1 for r in allr if r['correct'] == ''))
print('non-empty explanation:', sum(1 for r in allr if r['explanation']))
print('pilot records:', len(pilot))
print('pilot has_exhibit:', sum(r['has_exhibit'] for r in pilot))
print('pilot topics:', dict(collections.Counter(r['topic'] for r in pilot)))
print('\n=== PILOT TABLE (20) ===')
print(f"{'exam':<5} {'qnum':>4} {'page':>5} {'topic':<15} {'exhibit':<8} {'expl':<5} correct")
for p in pilot:
print(f"{p['exam']:<5} {p['qnum']:>4} {p['page']:>5} {p['topic']:<15} {str(p['has_exhibit']):<8} {str(bool(p['explanation'])):<5} {p['correct']}")

Xet Storage Details

Size:
1 kB
·
Xet hash:
95726f1d068d8f433c67fad3a4d276ad0445acfe10a67a7027e427c16510f7a4

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.