Training in progress epoch 40

README.md

@@ -14,24 +14,24 @@ 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: 0.1809
-- Validation Loss: 0.6326
-- Validation Mean Iou: 0.3331
-- Validation Mean Accuracy: 0.4129
-- Validation Overall Accuracy: 0.8566
-- Validation Per Category Iou: [0.         0.75063837 0.86934939 0.50549019 0.55209939 0.34707223
-        nan 0.47793191 0.53247141 0.10026632 0.83361171 0.00349941
- 0.         0.         0.01418999 0.58381467 0.         0.
- 0.72707071 0.09868955 0.47826501 0.33864732 0.05377358        nan
- 0.         0.39034504 0.23355263 0.         0.86185138 0.83043125
- 0.92062586 0.00839127 0.19885764 0.28062092 0.        ]
-- Validation Per Category Accuracy: [0.         0.8441298  0.95747519 0.60339777 0.66980687 0.43263595
-        nan 0.65140571 0.72588258 0.1153148  0.92360505 0.00462854
- 0.                nan 0.01448109 0.70674758 0.         0.
- 0.90597464 0.1045428  0.61409389 0.40161341 0.05444126        nan
- 0.         0.4704548  0.31055468 0.         0.94873549 0.91742269
- 0.9695204  0.01226634 0.4993465  0.35417153 0.        ]
-- Epoch: 39
+- Train Loss: 0.1724
+- Validation Loss: 0.6214
+- Validation Mean Iou: 0.3480
+- Validation Mean Accuracy: 0.4171
+- Validation Overall Accuracy: 0.8578
+- Validation Per Category Iou: [0.         0.74880719 0.86593712 0.49493695 0.5518544  0.34969174
+        nan 0.49195896 0.52033907 0.11512718 0.82580176 0.00328132
+ 0.                nan 0.         0.59486104 0.         0.
+ 0.73993215 0.15080459 0.49618358 0.37658279 0.03053435        nan
+ 0.         0.3952725  0.24578942 0.         0.85908382 0.81951817
+ 0.92798658 0.0131274  0.21907077 0.30066306 0.        ]
+- Validation Per Category Accuracy: [0.         0.83828969 0.95996416 0.58419522 0.67456154 0.43248759
+        nan 0.64852862 0.73357258 0.13170732 0.93534151 0.00384404
+ 0.                nan 0.         0.71573769 0.         0.
+ 0.88939744 0.16873945 0.67884542 0.42818992 0.03056351        nan
+ 0.         0.47664535 0.31638669 0.         0.94469242 0.92889975
+ 0.96830227 0.01681728 0.47420376 0.36689909 0.        ]
+- Epoch: 40
 
 ## Model description
 
@@ -563,6 +563,17 @@ The following hyperparameters were used during training:
  0.90597464 0.1045428  0.61409389 0.40161341 0.05444126        nan
  0.         0.4704548  0.31055468 0.         0.94873549 0.91742269
  0.9695204  0.01226634 0.4993465  0.35417153 0.        ]                                                                                                                                                        | 39    |
+| 0.1724     | 0.6214          | 0.3480              | 0.4171                   | 0.8578                      | [0.         0.74880719 0.86593712 0.49493695 0.5518544  0.34969174
+        nan 0.49195896 0.52033907 0.11512718 0.82580176 0.00328132
+ 0.                nan 0.         0.59486104 0.         0.
+ 0.73993215 0.15080459 0.49618358 0.37658279 0.03053435        nan
+ 0.         0.3952725  0.24578942 0.         0.85908382 0.81951817
+ 0.92798658 0.0131274  0.21907077 0.30066306 0.        ]                                                                                                                                                        | [0.         0.83828969 0.95996416 0.58419522 0.67456154 0.43248759
+        nan 0.64852862 0.73357258 0.13170732 0.93534151 0.00384404
+ 0.                nan 0.         0.71573769 0.         0.
+ 0.88939744 0.16873945 0.67884542 0.42818992 0.03056351        nan
+ 0.         0.47664535 0.31638669 0.         0.94469242 0.92889975
+ 0.96830227 0.01681728 0.47420376 0.36689909 0.        ]                                                                                                                                                        | 40    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40a7a6dfd8ff6ccebf36c89878a5dbb9bc2bbbc890501ad11b7ed987ee6da03d
+oid sha256:4ec8300d7156a8054c0f6c5417e77a1eae0de25907a5fcefb55a779d95172ea8
 size 15167588