[torchlib] Fix and improve quantization support #7430
| Job | Run time |
|---|---|
| 2m 48s | |
| 9m 25s | |
| 9m 10s | |
| 8m 13s | |
| 9m 40s | |
| 9m 1s | |
| 8m 47s | |
| 2m 23s | |
| 8m 59s | |
| 19m 5s | |
| 4s | |
| 2m 21s | |
| 3m 8s | |
| 17m 52s | |
| 15m 24s | |
| 21m 12s | |
| 17m 37s | |
| 3m 14s | |
| 11m 5s | |
| 2m 36s | |
| 3m 33s | |
| 10m 6s | |
| 7m 54s | |
| 9m 39s | |
| 9m 1s | |
| 1m 53s | |
| 6m 52s | |
| 1m 49s | |
| 1m 49s | |
| 4m 33s | |
| 3m 45s | |
| 5m 58s | |
| 6m 32s | |
| 25s | |
| 4h 15m 53s |