Training in progress epoch 36

README.md

@@ -14,30 +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: 0.1902
-- Validation Loss: 0.6752
-- Validation Mean Iou: 0.3282
-- Validation Mean Accuracy: 0.4101
-- Validation Overall Accuracy: 0.8514
-- Validation Per Category Iou: [0.00000000e+00 7.48854979e-01 8.61331067e-01 5.19863180e-01
- 5.79477385e-01 3.51250983e-01            nan 4.50917183e-01
- 5.23643337e-01 1.47324266e-01 8.00380433e-01 6.86907542e-04
- 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.47400472e-01
- 0.00000000e+00 0.00000000e+00 6.98699714e-01 4.83279925e-02
- 4.41216481e-01 3.86371527e-01 0.00000000e+00            nan
- 1.84009568e-04 3.73099406e-01 2.32542886e-01 0.00000000e+00
- 8.60124125e-01 8.38582887e-01 9.28199683e-01 1.04543843e-02
- 2.14406806e-01 2.68858654e-01 0.00000000e+00]
-- Validation Per Category Accuracy: [0.00000000e+00 8.10393846e-01 9.59458754e-01 6.49684677e-01
- 6.92065071e-01 4.67204091e-01            nan 6.05073291e-01
- 6.83021903e-01 1.97220647e-01 8.73930509e-01 7.84498313e-04
- 0.00000000e+00            nan 0.00000000e+00 7.49750275e-01
- 0.00000000e+00 0.00000000e+00 8.76772268e-01 5.11754773e-02
- 6.34910631e-01 4.72221966e-01 0.00000000e+00            nan
- 2.80347631e-04 4.48177854e-01 2.92171157e-01 0.00000000e+00
- 9.44228324e-01 9.22107275e-01 9.61649098e-01 1.58712035e-02
- 4.69801197e-01 3.44952309e-01 0.00000000e+00]
-- Epoch: 35
+- Train Loss: 0.1798
+- Validation Loss: 0.6439
+- Validation Mean Iou: 0.3460
+- Validation Mean Accuracy: 0.4195
+- Validation Overall Accuracy: 0.8508
+- Validation Per Category Iou: [0.         0.73276012 0.8703165  0.51981554 0.53793981 0.33233419
+        nan 0.45652634 0.52909238 0.15852233 0.81964961 0.03700225
+ 0.                nan 0.         0.57771173 0.         0.
+ 0.73584172 0.15695205 0.4595907  0.37447524 0.02385496        nan
+ 0.         0.36048183 0.26601379 0.         0.86087389 0.82180707
+ 0.92369107 0.00845113 0.20893951 0.29856993 0.        ]
+- Validation Per Category Accuracy: [0.         0.82418646 0.94532887 0.64936378 0.67010705 0.47232691
+        nan 0.61045369 0.68649566 0.19228588 0.92529247 0.04636385
+ 0.                nan 0.         0.7693662  0.         0.
+ 0.85725086 0.16617418 0.66391847 0.46183028 0.02387775        nan
+ 0.         0.41703531 0.33749604 0.         0.95559462 0.91895767
+ 0.97172331 0.02834959 0.45050561 0.3798439  0.        ]
+- Epoch: 36
 
 ## Model description
 
@@ -525,6 +519,17 @@ The following hyperparameters were used during training:
  2.80347631e-04 4.48177854e-01 2.92171157e-01 0.00000000e+00
  9.44228324e-01 9.22107275e-01 9.61649098e-01 1.58712035e-02
  4.69801197e-01 3.44952309e-01 0.00000000e+00] | 35    |
+| 0.1798     | 0.6439          | 0.3460              | 0.4195                   | 0.8508                      | [0.         0.73276012 0.8703165  0.51981554 0.53793981 0.33233419
+        nan 0.45652634 0.52909238 0.15852233 0.81964961 0.03700225
+ 0.                nan 0.         0.57771173 0.         0.
+ 0.73584172 0.15695205 0.4595907  0.37447524 0.02385496        nan
+ 0.         0.36048183 0.26601379 0.         0.86087389 0.82180707
+ 0.92369107 0.00845113 0.20893951 0.29856993 0.        ]                                                                                                                                                        | [0.         0.82418646 0.94532887 0.64936378 0.67010705 0.47232691
+        nan 0.61045369 0.68649566 0.19228588 0.92529247 0.04636385
+ 0.                nan 0.         0.7693662  0.         0.
+ 0.85725086 0.16617418 0.66391847 0.46183028 0.02387775        nan
+ 0.         0.41703531 0.33749604 0.         0.95559462 0.91895767
+ 0.97172331 0.02834959 0.45050561 0.3798439  0.        ]                                                                                                                                                        | 36    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b96e95f2d84a3c33f054f743b36c3ff2800065e8fe05792736fd2d0fdd6cbb92
+oid sha256:67a05066c192642fd18d655569aa9699553d479f0efbce1d3d00ece60cc2738d
 size 15167588