cs249r_book

mirror of https://github.com/harvard-edge/cs249r_book.git synced 2026-07-15 21:28:33 -05:00

Files

Rocky c7021f29cb fix(mlperf-edu): stop silently bypassing anti-cheat when mlperf CLI is missing (#1933 )

_process_zip() ran `mlperf verify` as the only anti-cheat check, then caught
FileNotFoundError and silently passed through to grading the submission's
self-reported JSON metrics verbatim with "Cheating": "NO" -- meaning any
environment without the mlperf CLI on PATH (the common case for a bare
grading checkout) accepted hand-forged submissions with no verification at
all.

Reproduced: built a zip with forged metrics (accuracy 0.99, throughput
99999) and confirmed _process_zip() returned Status: VALIDATED,
Cheating: NO. Now it returns Status: VERIFY_UNAVAILABLE, Cheating:
UNVERIFIED and logs the failure instead of grading blind.

2026-07-15 11:47:18 +02:00

autograder

fix(mlperf-edu): stop silently bypassing anti-cheat when mlperf CLI is missing (#1933 )

2026-07-15 11:47:18 +02:00

orchestration

Merge pull request #1938 from Shashank-Tripathi-07/fix/auto-trainer-plateau-history-reset

2026-07-10 23:43:08 +02:00

compliance_checker.py

Merge pull request #1931 from Shashank-Tripathi-07/fix/compliance-quality-target-lookup

2026-07-10 23:43:19 +02:00

generate_all_curves_v2.py

fix(mlperf-edu): use script-relative paths for paper figure outputs