Training in progress epoch 23

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: 0.2327
-- Validation Loss: 0.4708
-- Validation Mean Iou: 0.3333
-- Validation Mean Accuracy: 0.3991
-- Validation Overall Accuracy: 0.8674
-- Validation Per Category Iou: [0.00000000e+00 8.01482811e-01 8.67112634e-01 7.26469941e-01
- 8.47789494e-01 4.26344060e-01            nan 4.59877772e-01
- 5.61767489e-01 3.08784808e-02 8.50980045e-01 8.21140639e-04
- 0.00000000e+00 5.56009812e-02 0.00000000e+00 5.19460186e-01
- 0.00000000e+00 0.00000000e+00 6.66718429e-01 1.00749376e-01
- 3.06011822e-01 4.73609191e-01 0.00000000e+00            nan
- 5.21670878e-02 2.74661980e-01 2.16300138e-01 0.00000000e+00
- 8.70513680e-01 7.58257933e-01 9.31855744e-01 4.02310154e-04
- 5.41590367e-02 1.45066810e-01 0.00000000e+00]
-- Validation Per Category Accuracy: [0.00000000e+00 8.96213795e-01 9.66916102e-01 7.73539637e-01
- 8.80813661e-01 4.61368313e-01            nan 6.05202376e-01
- 7.84196522e-01 3.49120433e-02 9.33452042e-01 1.09083697e-03
- 0.00000000e+00 5.61056106e-02 0.00000000e+00 6.42177901e-01
- 0.00000000e+00 0.00000000e+00 7.70254403e-01 1.04428195e-01
- 7.01999428e-01 6.97985197e-01 0.00000000e+00            nan
- 5.98940589e-02 3.48124479e-01 2.34748471e-01 0.00000000e+00
- 9.37775670e-01 8.78224080e-01 9.72937862e-01 4.32992071e-04
- 2.44305382e-01 1.83551218e-01 0.00000000e+00]
-- Epoch: 22
+- Train Loss: 0.2307
+- Validation Loss: 0.4395
+- Validation Mean Iou: 0.3472
+- Validation Mean Accuracy: 0.4131
+- Validation Overall Accuracy: 0.8740
+- Validation Per Category Iou: [0.         0.78921013 0.87836164 0.7238651  0.85405051 0.48305222
+        nan 0.46174517 0.55335413 0.01711339 0.84971971 0.07427615
+ 0.         0.19647651 0.         0.52684296 0.         0.
+ 0.70735442 0.11184526 0.39826268 0.50387815 0.                nan
+ 0.05343915 0.27936942 0.23151827 0.         0.87119512 0.76032244
+ 0.93287485 0.00244547 0.02821955 0.16774339 0.        ]
+- Validation Per Category Accuracy: [0.         0.92820839 0.93256894 0.76932845 0.93401204 0.61981657
+        nan 0.61634072 0.81936678 0.01975643 0.94590099 0.1012495
+ 0.         0.2415429  0.         0.6241269  0.         0.
+ 0.81714715 0.1280285  0.64358053 0.67924551 0.                nan
+ 0.06437787 0.34337226 0.26928155 0.         0.94337279 0.8780219
+ 0.97322357 0.00300208 0.10171047 0.23578672 0.        ]
+- Epoch: 23
 
 ## Model description
 
@@ -370,6 +364,17 @@ The following hyperparameters were used during training:
  5.98940589e-02 3.48124479e-01 2.34748471e-01 0.00000000e+00
  9.37775670e-01 8.78224080e-01 9.72937862e-01 4.32992071e-04
  2.44305382e-01 1.83551218e-01 0.00000000e+00] | 22    |
+| 0.2307     | 0.4395          | 0.3472              | 0.4131                   | 0.8740                      | [0.         0.78921013 0.87836164 0.7238651  0.85405051 0.48305222
+        nan 0.46174517 0.55335413 0.01711339 0.84971971 0.07427615
+ 0.         0.19647651 0.         0.52684296 0.         0.
+ 0.70735442 0.11184526 0.39826268 0.50387815 0.                nan
+ 0.05343915 0.27936942 0.23151827 0.         0.87119512 0.76032244
+ 0.93287485 0.00244547 0.02821955 0.16774339 0.        ]                                                                                                                                                        | [0.         0.92820839 0.93256894 0.76932845 0.93401204 0.61981657
+        nan 0.61634072 0.81936678 0.01975643 0.94590099 0.1012495
+ 0.         0.2415429  0.         0.6241269  0.         0.
+ 0.81714715 0.1280285  0.64358053 0.67924551 0.                nan
+ 0.06437787 0.34337226 0.26928155 0.         0.94337279 0.8780219
+ 0.97322357 0.00300208 0.10171047 0.23578672 0.        ]                                                                                                                                                          | 23    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ade83caef35a6b0354879fa26b0d5da5e42568643ee8d76477ac4529cd300dd
+oid sha256:49f0c4de2ee118f00795d67401cbe3cbb4a51f1a5c11f58822e6156a14295ac0
 size 15167588