Commit 6206073
authored
[BUG] Distance calculations sporadically fail on sse. (#5500)
## Description of changes
The distance_sse functions started flaking in CI. It looks like they
are correct, just not the same stability as the baseline. Rather than
play a game of numerical stability to fix, I'm making the tolerance
proportional to vector length and thus the error. Also, up the number
of iterations so that failures are >50% likely rather than 1 in 100.
## Test plan
CI
## Migration plan
N/A
## Observability plan
N/A
## Documentation Changes
N/A1 parent adf862d commit 6206073
1 file changed
+10
-8
lines changed| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
361 | 361 | | |
362 | 362 | | |
363 | 363 | | |
364 | | - | |
| 364 | + | |
365 | 365 | | |
366 | 366 | | |
367 | 367 | | |
| |||
373 | 373 | | |
374 | 374 | | |
375 | 375 | | |
376 | | - | |
| 376 | + | |
377 | 377 | | |
378 | 378 | | |
379 | 379 | | |
| |||
385 | 385 | | |
386 | 386 | | |
387 | 387 | | |
388 | | - | |
| 388 | + | |
389 | 389 | | |
390 | 390 | | |
391 | 391 | | |
| |||
440 | 440 | | |
441 | 441 | | |
442 | 442 | | |
443 | | - | |
444 | | - | |
445 | | - | |
446 | | - | |
447 | | - | |
| 443 | + | |
| 444 | + | |
| 445 | + | |
| 446 | + | |
| 447 | + | |
| 448 | + | |
| 449 | + | |
448 | 450 | | |
449 | 451 | | |
450 | 452 | | |
| |||
0 commit comments