Training in progress epoch 38

README.md

@@ -15,13 +15,13 @@ 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: 0.0521
-- Train Dice Coef: 0.8503
-- Train Iou: 0.7481
-- Validation Loss: 0.0459
-- Validation Dice Coef: 0.8832
-- Validation Iou: 0.7919
+- Train Dice Coef: 0.8370
+- Train Iou: 0.7344
+- Validation Loss: 0.0451
+- Validation Dice Coef: 0.8845
+- Validation Iou: 0.7939
 - Train Lr: 1.25e-08
-- Epoch: 37
+- Epoch: 38
 
 ## Model description
 
@@ -85,6 +85,7 @@ The following hyperparameters were used during training:
 | 0.0522     | 0.8408          | 0.7385    | 0.0456          | 0.8838               | 0.7927         | 2.5e-07  | 35    |
 | 0.0521     | 0.8484          | 0.7454    | 0.0453          | 0.8839               | 0.7929         | 2.5e-07  | 36    |
 | 0.0521     | 0.8503          | 0.7481    | 0.0459          | 0.8832               | 0.7919         | 1.25e-08 | 37    |
+| 0.0521     | 0.8370          | 0.7344    | 0.0451          | 0.8845               | 0.7939         | 1.25e-08 | 38    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc1e1e9f0c5a6129be3cbdbd669099e035908e01f0d48c1049910bc20b532926
+oid sha256:a358755b8d1c9deda055755a8a70ebd71690beb3af0411df06325725c6d1e715
 size 15133664