Training in progress epoch 14

README.md

@@ -14,24 +14,30 @@ 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: 0.3146
-- Validation Loss: 0.4175
-- Validation Mean Iou: 0.3339
-- Validation Mean Accuracy: 0.3995
-- Validation Overall Accuracy: 0.8745
-- Validation Per Category Iou: [0.         0.81054591 0.88286867 0.68551149 0.86089895 0.4562385
-        nan 0.4522713  0.55496016 0.01456189 0.83576109 0.
- 0.         0.         0.         0.50709788 0.         0.
- 0.73464008 0.00175153 0.35021502 0.57263292 0.                nan
- 0.         0.25185222 0.14419755 0.         0.85952374 0.70281003
- 0.9270307  0.17660456 0.04867831 0.18762581 0.        ]
-- Validation Per Category Accuracy: [0.         0.9092016  0.94168672 0.86545289 0.89611216 0.55273728
-        nan 0.61409823 0.76682349 0.01569689 0.92776282 0.
- 0.         0.         0.         0.59972229 0.         0.
- 0.86700656 0.00175747 0.54181633 0.67419762 0.                nan
- 0.         0.3252672  0.14789466 0.         0.9316378  0.88743565
- 0.97060047 0.33277846 0.15319149 0.25967892 0.        ]
-- Epoch: 13
+- Train Loss: 0.3000
+- Validation Loss: 0.4196
+- Validation Mean Iou: 0.3263
+- Validation Mean Accuracy: 0.3833
+- Validation Overall Accuracy: 0.8720
+- Validation Per Category Iou: [0.00000000e+00 8.02547730e-01 8.74182776e-01 6.55641045e-01
+ 8.69918767e-01 4.12920686e-01            nan 4.34054109e-01
+ 5.54604573e-01 3.14830157e-03 8.29634841e-01 0.00000000e+00
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 4.98619437e-01
+ 0.00000000e+00 0.00000000e+00 7.20371619e-01 1.62799781e-02
+ 3.73295478e-01 5.20323501e-01 0.00000000e+00            nan
+ 3.48000087e-04 2.41829304e-01 1.50045164e-01 0.00000000e+00
+ 8.67415087e-01 7.31957881e-01 9.29791719e-01 1.28032094e-01
+ 2.77808135e-02 1.25956544e-01 0.00000000e+00]
+- Validation Per Category Accuracy: [0.00000000e+00 9.10809038e-01 9.53614030e-01 6.91330346e-01
+ 9.25106631e-01 4.73740259e-01            nan 5.64222160e-01
+ 7.49045544e-01 3.42805593e-03 9.38335743e-01 0.00000000e+00
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.77484642e-01
+ 0.00000000e+00 0.00000000e+00 8.68434883e-01 1.63507406e-02
+ 5.76763406e-01 7.07811962e-01 0.00000000e+00            nan
+ 3.51671539e-04 3.02660657e-01 1.55815731e-01 0.00000000e+00
+ 9.39832349e-01 8.43146236e-01 9.70195728e-01 2.11579170e-01
+ 1.06049228e-01 1.61502816e-01 0.00000000e+00]
+- Epoch: 14
 
 ## Model description
 
@@ -229,6 +235,23 @@ The following hyperparameters were used during training:
  0.86700656 0.00175747 0.54181633 0.67419762 0.                nan
  0.         0.3252672  0.14789466 0.         0.9316378  0.88743565
  0.97060047 0.33277846 0.15319149 0.25967892 0.        ]                                                                                                                                                                | 13    |
+| 0.3000     | 0.4196          | 0.3263              | 0.3833                   | 0.8720                      | [0.00000000e+00 8.02547730e-01 8.74182776e-01 6.55641045e-01
+ 8.69918767e-01 4.12920686e-01            nan 4.34054109e-01
+ 5.54604573e-01 3.14830157e-03 8.29634841e-01 0.00000000e+00
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 4.98619437e-01
+ 0.00000000e+00 0.00000000e+00 7.20371619e-01 1.62799781e-02
+ 3.73295478e-01 5.20323501e-01 0.00000000e+00            nan
+ 3.48000087e-04 2.41829304e-01 1.50045164e-01 0.00000000e+00
+ 8.67415087e-01 7.31957881e-01 9.29791719e-01 1.28032094e-01
+ 2.77808135e-02 1.25956544e-01 0.00000000e+00] | [0.00000000e+00 9.10809038e-01 9.53614030e-01 6.91330346e-01
+ 9.25106631e-01 4.73740259e-01            nan 5.64222160e-01
+ 7.49045544e-01 3.42805593e-03 9.38335743e-01 0.00000000e+00
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.77484642e-01
+ 0.00000000e+00 0.00000000e+00 8.68434883e-01 1.63507406e-02
+ 5.76763406e-01 7.07811962e-01 0.00000000e+00            nan
+ 3.51671539e-04 3.02660657e-01 1.55815731e-01 0.00000000e+00
+ 9.39832349e-01 8.43146236e-01 9.70195728e-01 2.11579170e-01
+ 1.06049228e-01 1.61502816e-01 0.00000000e+00] | 14    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3caaecd9ad82aba4cca0c46592605db39abe1893254d6be66380e1e659c22e3
+oid sha256:96cf625dd932a6eb6572da27cb7aacec0f0dcdd0ad676456a42c92de61f9ac29
 size 15167588