Training in progress epoch 28

README.md

@@ -14,14 +14,14 @@ probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Train Loss: 0.0192
-- Train Dice Coef: 0.7729
-- Train Iou: 0.6440
-- Validation Loss: 0.0234
-- Validation Dice Coef: 0.8483
-- Validation Iou: 0.7414
+- Train Loss: 0.0187
+- Train Dice Coef: 0.7849
+- Train Iou: 0.6572
+- Validation Loss: 0.0241
+- Validation Dice Coef: 0.8478
+- Validation Iou: 0.7398
 - Train Lr: 2.5e-07
-- Epoch: 27
+- Epoch: 28
 
 ## Model description
 
@@ -75,6 +75,7 @@ The following hyperparameters were used during training:
 | 0.0191     | 0.7761          | 0.6466    | 0.0238          | 0.8493               | 0.7419         | 5e-06    | 25    |
 | 0.0188     | 0.7781          | 0.6503    | 0.0237          | 0.8481               | 0.7405         | 5e-06    | 26    |
 | 0.0192     | 0.7729          | 0.6440    | 0.0234          | 0.8483               | 0.7414         | 2.5e-07  | 27    |
+| 0.0187     | 0.7849          | 0.6572    | 0.0241          | 0.8478               | 0.7398         | 2.5e-07  | 28    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7018d0b49ea400c6151964690c5b84d410befa472b535f537f4c1c98d5d9a6b
+oid sha256:309474b5bcff64a7a9f1a6fcc83e6c042504b8d844e5eef3ec5eb016ca61ee57
 size 15133664