curl -fsSLO https://puodziukas.dev/verify.sh && bash verify.sh --dry
curl -fsSLO https://puodziukas.dev/verify.sh && bash verify.sh| Claim | Command | Expected output | Links |
|---|---|---|---|
| 171/177 synthetic | python3 build_corpus.py && python3 evaluate.py | [171 / 177 adversarial caught] | receipt · repo · repo CI |
| 4/107 synthetic | python3 build_corpus.py && python3 evaluate.py | [4 benign over-blocked] | receipt · repo · repo CI |
| n=284 synthetic | python3 build_corpus.py && python3 evaluate.py | total=284) | receipt · repo · repo CI |
| $40,832.41 synthetic | python3 probe.py | Compounding error (loss): $40,832.40790 | receipt · repo · repo CI |
| 47,312 synthetic | python3 probe.py | Records processed : 47,312 | receipt · repo · repo CI |
| 6/177 synthetic | python3 build_corpus.py && python3 evaluate.py | [6 adversarial missed] | receipt · repo · repo CI |
| 0.96 measured-once | grep -F "Recall 0.96 on the in-house set" README.md | Recall 0.96 on the in-house set | receipt |
What a miss means
A MISS line is a claim that failed to reproduce, and the owner will cut it from the site.
