Training in progress epoch 31

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: 0.0225
-- Train Dice Coef: 0.7712
-- Train Iou: 0.6380
-- Validation Loss: 0.0245
-- Validation Dice Coef: 0.8496
-- Validation Iou: 0.7413
-- Train Lr: 1e-04
-- Epoch: 30
+- Train Loss: 0.0190
+- Train Dice Coef: 0.7867
+- Train Iou: 0.6579
+- Validation Loss: 0.0232
+- Validation Dice Coef: 0.8546
+- Validation Iou: 0.7486
+- Train Lr: 5e-06
+- Epoch: 31
 
 ## Model description
 
@@ -78,6 +78,7 @@ The following hyperparameters were used during training:
 | 0.0179     | 0.7973          | 0.6709    | 0.0247          | 0.8559               | 0.7504         | 1e-04    | 28    |
 | 0.0176     | 0.7959          | 0.6702    | 0.0245          | 0.8567               | 0.7515         | 1e-04    | 29    |
 | 0.0225     | 0.7712          | 0.6380    | 0.0245          | 0.8496               | 0.7413         | 1e-04    | 30    |
+| 0.0190     | 0.7867          | 0.6579    | 0.0232          | 0.8546               | 0.7486         | 5e-06    | 31    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d443d368335aeb3b4e906905c0d09b32934d79f1311d55bb4dff8c3cb75562e1
+oid sha256:931ad5e301bddf8f8888f8d759b4668a6ff3421906ce3bdb9b56e94f7677fd8d
 size 15133664