Training in progress epoch 69

README.md

@@ -16,12 +16,12 @@ This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvi
 It achieves the following results on the evaluation set:
 - Train Loss: 0.0161
 - Train Dice Coef: 0.8047
-- Train Iou: 0.6812
-- Validation Loss: 0.0226
-- Validation Dice Coef: 0.8566
-- Validation Iou: 0.7513
+- Train Iou: 0.6813
+- Validation Loss: 0.0230
+- Validation Dice Coef: 0.8564
+- Validation Iou: 0.7510
 - Train Lr: 1e-10
-- Epoch: 68
+- Epoch: 69
 
 ## Model description
 
@@ -116,6 +116,7 @@ The following hyperparameters were used during training:
 | 0.0161     | 0.8030          | 0.6795    | 0.0226          | 0.8570               | 0.7520         | 1e-10    | 66    |
 | 0.0163     | 0.8046          | 0.6813    | 0.0223          | 0.8579               | 0.7534         | 1e-10    | 67    |
 | 0.0161     | 0.8047          | 0.6812    | 0.0226          | 0.8566               | 0.7513         | 1e-10    | 68    |
+| 0.0161     | 0.8047          | 0.6813    | 0.0230          | 0.8564               | 0.7510         | 1e-10    | 69    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1b14c369568a39cba4ec809d60a02fcca6adcdfabc6c59abc9818e68b7ea613b
+oid sha256:b43f614ce83561d7c573739e539fa69c9d903cb454a13acd401cd7709aac1f04
 size 15133664