Training in progress epoch 40

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: 0.0164
-- Train Dice Coef: 0.7995
-- Train Iou: 0.6753
-- Validation Loss: 0.0225
-- Validation Dice Coef: 0.8568
-- Validation Iou: 0.7516
-- Train Lr: 5e-06
-- Epoch: 39
+- Train Loss: 0.0160
+- Train Dice Coef: 0.8077
+- Train Iou: 0.6853
+- Validation Loss: 0.0229
+- Validation Dice Coef: 0.8559
+- Validation Iou: 0.7503
+- Train Lr: 2.5e-07
+- Epoch: 40
 
 ## Model description
 
@@ -87,6 +87,7 @@ The following hyperparameters were used during training:
 | 0.0167     | 0.8025          | 0.6786    | 0.0226          | 0.8561               | 0.7507         | 5e-06    | 37    |
 | 0.0163     | 0.8027          | 0.6791    | 0.0231          | 0.8560               | 0.7505         | 5e-06    | 38    |
 | 0.0164     | 0.7995          | 0.6753    | 0.0225          | 0.8568               | 0.7516         | 5e-06    | 39    |
+| 0.0160     | 0.8077          | 0.6853    | 0.0229          | 0.8559               | 0.7503         | 2.5e-07  | 40    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8fa1a442f3639fdbbbf00a88c49f5993798295b70f7493c6242d30f24ce2c488
+oid sha256:6ccb1172e0606e17dcde9c0e119b934c07d9817b54dcdabdbd4170e8f27b4635
 size 15133664