Model
ResNet50 v1.5 MXNet checkpoint trained on Imagenet with AMP
Use the NGC CLI to download:
Copied!
1 Version
20.12.0_ampSelected09/22/2022 6:03 PM UTC49.08 MBAccuracy: 0.771290 EpochsBatch Size: 192GPU: V100 Copied!
20.12.0_ampSelected
09/22/2022 6:03 PM UTC49.08 MBAccuracy: 0.771290 EpochsBatch Size: 192GPU: V100
Copied!
architecture
| Key | Value |
|---|---|
| task | classification |
| type | rn50 |
performance
| Key | Value |
|---|---|
| accuracy | 77.12% |
| training_loss | 1.745 |
training
| Key | Value |
|---|---|
| batch_size | 192 |
| training_precision | AMP |
| dataset | ImageNet |
| epochs | 90 |