Model
UNet_Medical TensorFlow2 checkpoint trained with AMP
Use the NGC CLI to download:
Copied!
1 Version
21.11.0Selected02/03/2022 10:15 PM UTC394.99 MBAccuracy: 00 EpochsBatch Size: 8GPU: A100 Copied!
21.11.0Selected
02/03/2022 10:15 PM UTC394.99 MBAccuracy: 00 EpochsBatch Size: 8GPU: A100
Copied!
architecture
| Key | Value |
|---|---|
| type | base |
performance
| Key | Value |
|---|---|
| eval_dice_loss | 0.09937584400177002 |
| eval_dice_score | 0.90062415599823 |
| eval_total_loss | 0.29639625549316406 |
| eval_ce_loss | 0.19702039659023285 |
training
| Key | Value |
|---|---|
| batch_size | 8 |
| training_precision | AMP |
| dataset | EM segmentation challenge |
| learning_rate | 0.0001 |
| iterations | 2000 |