Training in progress epoch 46

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: 0.1781
-- Validation Loss: 0.4529
-- Validation Mean Iou: 0.3574
-- Validation Mean Accuracy: 0.4191
-- Validation Overall Accuracy: 0.8774
-- Validation Per Category Iou: [0.         0.80590718 0.86988106 0.71453937 0.86614986 0.4249264
-        nan 0.47975456 0.54781776 0.09453882 0.84344987 0.10822898
- 0.         0.05507559 0.         0.52010855 0.         0.
- 0.74685485 0.08247778 0.43946981 0.52154191 0.24444444        nan
- 0.03570312 0.29856758 0.28932541 0.         0.86943451 0.76210843
- 0.93975656 0.04628113 0.00666137 0.17955183 0.        ]
-- Validation Per Category Accuracy: [0.         0.90777016 0.93538454 0.74532453 0.96011721 0.53476928
-        nan 0.67329934 0.81065134 0.10744249 0.95209245 0.16798889
- 0.         0.06311881 0.         0.58061937 0.         0.
- 0.86390726 0.09013307 0.63941956 0.64314678 0.28530259        nan
- 0.04266215 0.38685357 0.33004447 0.         0.94754825 0.87771988
- 0.97640668 0.04758102 0.0090947  0.25304287 0.        ]
-- Epoch: 45
+- Train Loss: 0.1684
+- Validation Loss: 0.4189
+- Validation Mean Iou: 0.3690
+- Validation Mean Accuracy: 0.4293
+- Validation Overall Accuracy: 0.8874
+- Validation Per Category Iou: [0.         0.84396057 0.88758961 0.81114547 0.88917721 0.45907522
+        nan 0.49766899 0.57141046 0.13114032 0.86950251 0.16812566
+ 0.         0.05402798 0.         0.55671436 0.         0.
+ 0.75187936 0.06411064 0.38664923 0.50159138 0.2393736         nan
+ 0.03612036 0.31872514 0.30305254 0.         0.86860485 0.76230745
+ 0.94340831 0.00953726 0.07691803 0.17518119 0.        ]
+- Validation Per Category Accuracy: [0.         0.91434158 0.94889397 0.87210345 0.95322397 0.59465787
+        nan 0.66608234 0.78205623 0.16111863 0.93973807 0.20378818
+ 0.         0.06930693 0.         0.66147017 0.         0.
+ 0.89656164 0.06706628 0.53578308 0.5492624  0.30835735        nan
+ 0.04147526 0.40549508 0.35870623 0.         0.94735747 0.87335684
+ 0.97257866 0.01137326 0.19582812 0.23648119 0.        ]
+- Epoch: 46
 
 ## Model description
 
@@ -629,6 +629,17 @@ The following hyperparameters were used during training:
  0.86390726 0.09013307 0.63941956 0.64314678 0.28530259        nan
  0.04266215 0.38685357 0.33004447 0.         0.94754825 0.87771988
  0.97640668 0.04758102 0.0090947  0.25304287 0.        ]                                                                                                                                                        | 45    |
+| 0.1684     | 0.4189          | 0.3690              | 0.4293                   | 0.8874                      | [0.         0.84396057 0.88758961 0.81114547 0.88917721 0.45907522
+        nan 0.49766899 0.57141046 0.13114032 0.86950251 0.16812566
+ 0.         0.05402798 0.         0.55671436 0.         0.
+ 0.75187936 0.06411064 0.38664923 0.50159138 0.2393736         nan
+ 0.03612036 0.31872514 0.30305254 0.         0.86860485 0.76230745
+ 0.94340831 0.00953726 0.07691803 0.17518119 0.        ]                                                                                                                                                        | [0.         0.91434158 0.94889397 0.87210345 0.95322397 0.59465787
+        nan 0.66608234 0.78205623 0.16111863 0.93973807 0.20378818
+ 0.         0.06930693 0.         0.66147017 0.         0.
+ 0.89656164 0.06706628 0.53578308 0.5492624  0.30835735        nan
+ 0.04147526 0.40549508 0.35870623 0.         0.94735747 0.87335684
+ 0.97257866 0.01137326 0.19582812 0.23648119 0.        ]                                                                                                                                                        | 46    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f36deb2506de683e9ef30a4d7ea112c537ce13c3cd1daf4e4053e37cea35c6a4
+oid sha256:d9a31581605396932616f36c8387b4ab98796cdc30e36f71747d37514a46f94c
 size 15167588