Training in progress epoch 43

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: 0.1886
-- Validation Loss: 0.4251
-- Validation Mean Iou: 0.3676
-- Validation Mean Accuracy: 0.4456
-- Validation Overall Accuracy: 0.8827
-- Validation Per Category Iou: [0.         0.81869055 0.88872516 0.83047488 0.84276443 0.50458334
-        nan 0.50154127 0.55514409 0.11767963 0.8593002  0.18142472
- 0.         0.17150681 0.         0.52765579 0.         0.
- 0.72523268 0.11768383 0.38784377 0.55537055 0.03125           nan
- 0.04156492 0.2927821  0.28836848 0.         0.87210633 0.77489312
- 0.94023927 0.01637101 0.07383087 0.2126436  0.        ]
-- Validation Per Category Accuracy: [0.         0.91039924 0.94692984 0.8782822  0.93049806 0.63547194
-        nan 0.64587233 0.81659018 0.1533604  0.95128203 0.25357001
- 0.         0.5554868  0.         0.59909114 0.         0.
- 0.81739862 0.14772157 0.70659589 0.64127558 0.03170029        nan
- 0.05387168 0.38205471 0.34779739 0.         0.9516963  0.85983713
- 0.97204659 0.02417058 0.18481435 0.30708968 0.        ]
-- Epoch: 42
+- Train Loss: 0.1649
+- Validation Loss: 0.4242
+- Validation Mean Iou: 0.3717
+- Validation Mean Accuracy: 0.4383
+- Validation Overall Accuracy: 0.8829
+- Validation Per Category Iou: [0.         0.8245252  0.88278    0.781006   0.85842353 0.51259623
+        nan 0.51245584 0.58046843 0.1180867  0.86495296 0.18057803
+ 0.         0.19290237 0.         0.57019005 0.         0.
+ 0.73953555 0.08630286 0.40811401 0.47419294 0.14364641        nan
+ 0.03056729 0.30467132 0.2893461  0.         0.86773127 0.73354145
+ 0.94299225 0.12116164 0.05836818 0.18823529 0.        ]
+- Validation Per Category Accuracy: [0.         0.89896041 0.94310935 0.79709015 0.94969977 0.67147032
+        nan 0.65138807 0.79877367 0.14253496 0.94740116 0.23234827
+ 0.         0.27021452 0.         0.66277455 0.         0.
+ 0.86192885 0.09854381 0.60905623 0.64588831 0.14985591        nan
+ 0.03674968 0.38943084 0.3317468  0.         0.93877576 0.88623506
+ 0.97329085 0.17635286 0.14092616 0.25900882 0.        ]
+- Epoch: 43
 
 ## Model description
 
@@ -596,6 +596,17 @@ The following hyperparameters were used during training:
  0.81739862 0.14772157 0.70659589 0.64127558 0.03170029        nan
  0.05387168 0.38205471 0.34779739 0.         0.9516963  0.85983713
  0.97204659 0.02417058 0.18481435 0.30708968 0.        ]                                                                                                                                                        | 42    |
+| 0.1649     | 0.4242          | 0.3717              | 0.4383                   | 0.8829                      | [0.         0.8245252  0.88278    0.781006   0.85842353 0.51259623
+        nan 0.51245584 0.58046843 0.1180867  0.86495296 0.18057803
+ 0.         0.19290237 0.         0.57019005 0.         0.
+ 0.73953555 0.08630286 0.40811401 0.47419294 0.14364641        nan
+ 0.03056729 0.30467132 0.2893461  0.         0.86773127 0.73354145
+ 0.94299225 0.12116164 0.05836818 0.18823529 0.        ]                                                                                                                                                        | [0.         0.89896041 0.94310935 0.79709015 0.94969977 0.67147032
+        nan 0.65138807 0.79877367 0.14253496 0.94740116 0.23234827
+ 0.         0.27021452 0.         0.66277455 0.         0.
+ 0.86192885 0.09854381 0.60905623 0.64588831 0.14985591        nan
+ 0.03674968 0.38943084 0.3317468  0.         0.93877576 0.88623506
+ 0.97329085 0.17635286 0.14092616 0.25900882 0.        ]                                                                                                                                                        | 43    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e24b69bc1bbed3a3eb0890c976bbd4ce42f391af7bd9e94af8a39f930d388ca6
+oid sha256:5d3ca32d742ae212d47aae37ec54615a34b691364d57e32f1ccb0b4b2b27c191
 size 15167588