Training in progress epoch 97

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: 0.0524
-- Train Dice Coef: 0.8417
-- Train Iou: 0.7398
-- Validation Loss: 0.0453
-- Validation Dice Coef: 0.8831
-- Validation Iou: 0.7917
+- Train Loss: 0.0517
+- Train Dice Coef: 0.8410
+- Train Iou: 0.7390
+- Validation Loss: 0.0440
+- Validation Dice Coef: 0.8870
+- Validation Iou: 0.7979
 - Train Lr: 1e-10
-- Epoch: 96
+- Epoch: 97
 
 ## Model description
 
@@ -144,6 +144,7 @@ The following hyperparameters were used during training:
 | 0.0525     | 0.8399          | 0.7371    | 0.0452          | 0.8846               | 0.7940         | 1e-10    | 94    |
 | 0.0517     | 0.8433          | 0.7401    | 0.0455          | 0.8835               | 0.7924         | 1e-10    | 95    |
 | 0.0524     | 0.8417          | 0.7398    | 0.0453          | 0.8831               | 0.7917         | 1e-10    | 96    |
+| 0.0517     | 0.8410          | 0.7390    | 0.0440          | 0.8870               | 0.7979         | 1e-10    | 97    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05c526b7dec908e714aeedc51f9fa03dfd79cfa9dc37cf6ebad5fc72035d8100
+oid sha256:e6fb895f2b50932b462bb18701145a30ea08733b5f9f20b5fd50a148a891cdbe
 size 15133664