AdamE optimizer with decoupled L1 and L2 regularization #595
| Job | Run time |
|---|---|
| 12s | |
| 8m 23s | |
| 14s | |
| 8m 22s | |
| 16s | |
| 9m 40s | |
| 1m 18s | |
| 9m 27s | |
| 11m 20s | |
| 10m 38s | |
| 9m 24s | |
| 10m 56s | |
| 9m 46s | |
| 10m 40s | |
| 12m 0s | |
| 12m 36s | |
| 13m 35s | |
| 13m 9s | |
| 12m 30s | |
| 12m 29s | |
| 2m 9s | |
| 16s | |
| 16s | |
| 2m 38s | |
| 14s | |
| 8s | |
| 3h 2m 36s |