Training in progress epoch 20

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: 0.2570
-- Validation Loss: 0.4381
-- Validation Mean Iou: 0.3378
-- Validation Mean Accuracy: 0.4040
-- Validation Overall Accuracy: 0.8691
-- Validation Per Category Iou: [0.         0.78633412 0.8781239  0.70951789 0.85768155 0.49725305
-        nan 0.4385802  0.5419402  0.01325455 0.84049064 0.03469167
- 0.         0.         0.         0.52032603 0.         0.
- 0.68820155 0.07929718 0.30712852 0.51640481 0.                nan
- 0.01769049 0.26803817 0.21887178 0.         0.85998636 0.71539146
- 0.93235425 0.24885785 0.05621853 0.11969413 0.        ]
-- Validation Per Category Accuracy: [0.         0.91321796 0.93586512 0.7493935  0.91472526 0.63834931
-        nan 0.58292224 0.81417994 0.01497519 0.94252235 0.05394685
- 0.         0.         0.         0.64331398 0.         0.
- 0.82029437 0.08115742 0.56811405 0.59644195 0.                nan
- 0.01995736 0.34179208 0.24586576 0.         0.94413845 0.83304234
- 0.96807676 0.34801978 0.20125156 0.15898892 0.        ]
-- Epoch: 19
+- Train Loss: 0.2617
+- Validation Loss: 0.4168
+- Validation Mean Iou: 0.3396
+- Validation Mean Accuracy: 0.3963
+- Validation Overall Accuracy: 0.8781
+- Validation Per Category Iou: [0.00000000e+00 7.94986290e-01 8.78321279e-01 7.49897343e-01
+ 8.49326301e-01 5.23130579e-01            nan 4.50929207e-01
+ 5.51662857e-01 2.18050542e-02 8.41160082e-01 1.61248710e-02
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 4.99800580e-01
+ 0.00000000e+00 0.00000000e+00 7.33030551e-01 3.70162822e-02
+ 3.87012787e-01 5.37036435e-01 0.00000000e+00            nan
+ 2.52828519e-04 2.58401363e-01 2.18729726e-01 0.00000000e+00
+ 8.68371051e-01 7.68056025e-01 9.33727233e-01 9.82409932e-02
+ 3.83513478e-02 1.51214616e-01 0.00000000e+00]
+- Validation Per Category Accuracy: [0.00000000e+00 9.07468689e-01 9.33071883e-01 8.06640187e-01
+ 9.49407168e-01 6.81786840e-01            nan 5.64532420e-01
+ 8.05049940e-01 2.72440235e-02 9.42797113e-01 2.47917493e-02
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.80009257e-01
+ 0.00000000e+00 0.00000000e+00 8.80850860e-01 3.73148381e-02
+ 5.59445628e-01 5.88173859e-01 0.00000000e+00            nan
+ 2.63753654e-04 3.24654954e-01 2.34262090e-01 0.00000000e+00
+ 9.43142842e-01 8.79414683e-01 9.68600549e-01 1.42839273e-01
+ 9.82895286e-02 1.98829554e-01 0.00000000e+00]
+- Epoch: 20
 
 ## Model description
 
@@ -313,6 +319,23 @@ The following hyperparameters were used during training:
  0.82029437 0.08115742 0.56811405 0.59644195 0.                nan
  0.01995736 0.34179208 0.24586576 0.         0.94413845 0.83304234
  0.96807676 0.34801978 0.20125156 0.15898892 0.        ]                                                                                                                                                        | 19    |
+| 0.2617     | 0.4168          | 0.3396              | 0.3963                   | 0.8781                      | [0.00000000e+00 7.94986290e-01 8.78321279e-01 7.49897343e-01
+ 8.49326301e-01 5.23130579e-01            nan 4.50929207e-01
+ 5.51662857e-01 2.18050542e-02 8.41160082e-01 1.61248710e-02
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 4.99800580e-01
+ 0.00000000e+00 0.00000000e+00 7.33030551e-01 3.70162822e-02
+ 3.87012787e-01 5.37036435e-01 0.00000000e+00            nan
+ 2.52828519e-04 2.58401363e-01 2.18729726e-01 0.00000000e+00
+ 8.68371051e-01 7.68056025e-01 9.33727233e-01 9.82409932e-02
+ 3.83513478e-02 1.51214616e-01 0.00000000e+00] | [0.00000000e+00 9.07468689e-01 9.33071883e-01 8.06640187e-01
+ 9.49407168e-01 6.81786840e-01            nan 5.64532420e-01
+ 8.05049940e-01 2.72440235e-02 9.42797113e-01 2.47917493e-02
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.80009257e-01
+ 0.00000000e+00 0.00000000e+00 8.80850860e-01 3.73148381e-02
+ 5.59445628e-01 5.88173859e-01 0.00000000e+00            nan
+ 2.63753654e-04 3.24654954e-01 2.34262090e-01 0.00000000e+00
+ 9.43142842e-01 8.79414683e-01 9.68600549e-01 1.42839273e-01
+ 9.82895286e-02 1.98829554e-01 0.00000000e+00] | 20    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f50edad3f048cf7b8e7b851a6e0bfd96f035a1125a0fd10bf56b09d7d7d2a1a1
+oid sha256:eac7b67c80e9430b96a394eefb865f61e21bb5febeb5798c4fd1ba51ef2cb3f0
 size 15167588