Training in progress epoch 44

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: 0.1747
-- Validation Loss: 0.6300
-- Validation Mean Iou: 0.3453
-- Validation Mean Accuracy: 0.4256
-- Validation Overall Accuracy: 0.8568
-- Validation Per Category Iou: [0.         0.73808492 0.87628034 0.52271507 0.56354951 0.37292345
-        nan 0.47493813 0.54315812 0.12941955 0.83487747 0.01042155
- 0.         0.         0.01888056 0.5850729  0.         0.00679561
- 0.72865655 0.16298805 0.50779295 0.35281309 0.1483209         nan
- 0.00146372 0.39135342 0.26493341 0.         0.85556625 0.80147756
- 0.92444109 0.02342468 0.24566577 0.30760098 0.        ]
-- Validation Per Category Accuracy: [0.         0.82415433 0.95253909 0.65333127 0.67595726 0.52431439
-        nan 0.64563739 0.70549967 0.15530346 0.92934642 0.01396407
- 0.                nan 0.01890587 0.67997703 0.         0.00720222
- 0.90355305 0.19458696 0.64474034 0.38600282 0.15186246        nan
- 0.00294365 0.45284562 0.34301109 0.         0.93293094 0.94313517
- 0.96948846 0.04120314 0.47275917 0.39519926 0.        ]
-- Epoch: 43
+- Train Loss: 0.1633
+- Validation Loss: 0.6958
+- Validation Mean Iou: 0.3460
+- Validation Mean Accuracy: 0.4328
+- Validation Overall Accuracy: 0.8496
+- Validation Per Category Iou: [0.00000000e+00 7.07967513e-01 8.64897292e-01 5.30492291e-01
+ 5.41351423e-01 3.52224043e-01            nan 4.81002598e-01
+ 5.30743896e-01 1.82704272e-01 8.36799496e-01 3.19967322e-03
+ 0.00000000e+00 0.00000000e+00 2.70414280e-02 6.07536390e-01
+ 0.00000000e+00 2.58297817e-04 7.24070745e-01 1.61121955e-01
+ 4.93941753e-01 3.39722827e-01 1.68744805e-01            nan
+ 2.13736107e-04 3.79707826e-01 2.58062945e-01 0.00000000e+00
+ 8.60635733e-01 8.34524347e-01 9.28107342e-01 3.52373795e-02
+ 2.69899565e-01 2.96337021e-01 0.00000000e+00]
+- Validation Per Category Accuracy: [0.00000000e+00 7.80104458e-01 9.38267660e-01 6.60722284e-01
+ 7.00715501e-01 5.74914256e-01            nan 6.46841568e-01
+ 6.89412555e-01 2.99829836e-01 9.37107209e-01 3.68714207e-03
+ 0.00000000e+00            nan 2.72190936e-02 7.24927580e-01
+ 0.00000000e+00 2.77008310e-04 9.00758420e-01 1.80932589e-01
+ 6.38642366e-01 3.89678886e-01 1.93887297e-01            nan
+ 4.20521447e-04 4.50414586e-01 3.35784469e-01 0.00000000e+00
+ 9.51069425e-01 8.99380209e-01 9.68441171e-01 6.96016703e-02
+ 4.95425466e-01 3.89915943e-01 0.00000000e+00]
+- Epoch: 44
 
 ## Model description
 
@@ -613,6 +619,23 @@ The following hyperparameters were used during training:
  0.90355305 0.19458696 0.64474034 0.38600282 0.15186246        nan
  0.00294365 0.45284562 0.34301109 0.         0.93293094 0.94313517
  0.96948846 0.04120314 0.47275917 0.39519926 0.        ]                                                                                                                                                | 43    |
+| 0.1633     | 0.6958          | 0.3460              | 0.4328                   | 0.8496                      | [0.00000000e+00 7.07967513e-01 8.64897292e-01 5.30492291e-01
+ 5.41351423e-01 3.52224043e-01            nan 4.81002598e-01
+ 5.30743896e-01 1.82704272e-01 8.36799496e-01 3.19967322e-03
+ 0.00000000e+00 0.00000000e+00 2.70414280e-02 6.07536390e-01
+ 0.00000000e+00 2.58297817e-04 7.24070745e-01 1.61121955e-01
+ 4.93941753e-01 3.39722827e-01 1.68744805e-01            nan
+ 2.13736107e-04 3.79707826e-01 2.58062945e-01 0.00000000e+00
+ 8.60635733e-01 8.34524347e-01 9.28107342e-01 3.52373795e-02
+ 2.69899565e-01 2.96337021e-01 0.00000000e+00] | [0.00000000e+00 7.80104458e-01 9.38267660e-01 6.60722284e-01
+ 7.00715501e-01 5.74914256e-01            nan 6.46841568e-01
+ 6.89412555e-01 2.99829836e-01 9.37107209e-01 3.68714207e-03
+ 0.00000000e+00            nan 2.72190936e-02 7.24927580e-01
+ 0.00000000e+00 2.77008310e-04 9.00758420e-01 1.80932589e-01
+ 6.38642366e-01 3.89678886e-01 1.93887297e-01            nan
+ 4.20521447e-04 4.50414586e-01 3.35784469e-01 0.00000000e+00
+ 9.51069425e-01 8.99380209e-01 9.68441171e-01 6.96016703e-02
+ 4.95425466e-01 3.89915943e-01 0.00000000e+00] | 44    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d2021121dbc09e082d91efdc6c295f430a8fd6ac18e989d84c73ea57b053fe4
+oid sha256:0a7366d97c1a9303b0b80c2446825a12aa2663608a8002a45fcd166347e94597
 size 15167588