ruihan/qwen38-flash-next-dgx-spark
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Record successful 2048-token CUDA Graph retest and preserve failure samples

Browse Source
This commit is contained in:
ruihan
2026-09-17 23:04:40 +08:00
parent 1e2f48d1a4
commit 48a1c9d7c4
19 changed files with 430 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files
scripts/summarize-benchmark.py
+4 -4
View File
@@ -14,13 +14,13 @@ for path in args.files:
rows.append(json.loads(line))
except ValueError:
continue
short = [r for r in rows if r.get('test', '').startswith('short_')]
short = [r for r in rows if r.get('test', '').startswith('short_') and 'correct' in r]
summary = {'file': path.name,
'completed': any(r.get('event') == 'BENCHMARK_PASS' for r in rows),
'completed': any(r.get('event') in ('BENCHMARK_PASS', 'BENCHMARK_FAILED', 'BENCHMARK_COMPLETED') for r in rows),
'short_decode_tps_median': statistics.median(r['decode_tps_approx'] for r in short) if short else None,
'short_ttft_s_median': statistics.median(r['ttft_s'] for r in short) if short else None,
'all_checked_answers_correct': all(r['correct'] for r in rows if 'correct' in r),
'long': [{k:r[k] for k in ['test','ttft_s','elapsed_s','correct']} for r in rows if r.get('test','').startswith('long_')],
'all_checked_answers_correct': bool([r for r in rows if 'correct' in r]) and all(r['correct'] and r.get('finish_reason') == 'stop' for r in rows if 'correct' in r),
'long': [{k:r[k] for k in ['test','ttft_s','elapsed_s','correct']} for r in rows if r.get('test','').startswith('long_') and 'correct' in r],
'concurrency': [r for r in rows if r.get('test') == 'concurrency_summary'],
'prefix': [r for r in rows if r.get('test','').startswith('prefix_')]}
print(json.dumps(summary, ensure_ascii=False, indent=2))