bugfix: avoid randomly generating negative indices in samples #905
| Job | Run time |
|---|---|
| 15m 18s | |
| 17m 50s | |
| 5m 38s | |
| 6m 48s | |
| 4m 38s | |
| 4m 28s | |
| 12m 36s | |
| 12m 21s | |
| 11m 32s | |
| 11m 46s | |
| 12m 21s | |
| 11m 14s | |
| 5m 31s | |
| 5m 4s | |
| 15m 19s | |
| 15m 20s | |
| 2h 47m 44s |
| Job | Run time |
|---|---|
| 15m 18s | |
| 17m 50s | |
| 5m 38s | |
| 6m 48s | |
| 4m 38s | |
| 4m 28s | |
| 12m 36s | |
| 12m 21s | |
| 11m 32s | |
| 11m 46s | |
| 12m 21s | |
| 11m 14s | |
| 5m 31s | |
| 5m 4s | |
| 15m 19s | |
| 15m 20s | |
| 2h 47m 44s |