Training in progress epoch 33

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: 0.1898
-- Validation Loss: 0.6369
-- Validation Mean Iou: 0.3434
-- Validation Mean Accuracy: 0.4111
-- Validation Overall Accuracy: 0.8572
-- Validation Per Category Iou: [0.         0.77005646 0.8691602  0.51421807 0.54505788 0.37509283
-        nan 0.47473811 0.5177013  0.10361726 0.82314535 0.
- 0.                nan 0.         0.55573429 0.         0.
- 0.72789094 0.16284579 0.46446138 0.32311816 0.                nan
- 0.         0.38446267 0.21404751 0.         0.86325521 0.8362079
- 0.92635894 0.01265831 0.22255217 0.30201657 0.        ]
-- Validation Per Category Accuracy: [0.         0.84783576 0.94994793 0.63546724 0.67852851 0.47478672
-        nan 0.65145622 0.67587993 0.11633579 0.93587159 0.
- 0.                nan 0.         0.6686395  0.         0.
- 0.90790966 0.17633784 0.62937796 0.34917531 0.                nan
- 0.         0.4742821  0.2644691  0.         0.94883563 0.91963887
- 0.96496596 0.03255799 0.46656807 0.38725811 0.        ]
-- Epoch: 32
+- Train Loss: 0.1922
+- Validation Loss: 0.6796
+- Validation Mean Iou: 0.3286
+- Validation Mean Accuracy: 0.4105
+- Validation Overall Accuracy: 0.8482
+- Validation Per Category Iou: [0.00000000e+00 7.40810462e-01 8.66118742e-01 4.99364852e-01
+ 5.26878756e-01 3.72552936e-01            nan 4.74841300e-01
+ 5.25079443e-01 1.14740050e-01 8.35714370e-01 6.38207346e-04
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.74621701e-01
+ 0.00000000e+00 0.00000000e+00 7.07495193e-01 1.41447126e-01
+ 4.62213084e-01 3.40642698e-01 0.00000000e+00            nan
+ 0.00000000e+00 3.89053890e-01 2.22898422e-01 0.00000000e+00
+ 8.45321982e-01 7.94221222e-01 9.17872097e-01 1.76532216e-02
+ 2.30290833e-01 2.42034008e-01 0.00000000e+00]
+- Validation Per Category Accuracy: [0.00000000e+00 8.16530181e-01 9.52002646e-01 6.63124748e-01
+ 6.26037619e-01 4.92956588e-01            nan 6.47597208e-01
+ 6.85001856e-01 1.36698809e-01 9.31706346e-01 7.06048482e-04
+ 0.00000000e+00            nan 0.00000000e+00 6.94635901e-01
+ 0.00000000e+00 0.00000000e+00 8.95976958e-01 1.87069749e-01
+ 5.59829035e-01 4.10310044e-01 0.00000000e+00            nan
+ 0.00000000e+00 4.87146698e-01 2.85578447e-01 0.00000000e+00
+ 9.25510256e-01 9.23843473e-01 9.69757916e-01 4.94731348e-02
+ 4.80154090e-01 3.14923397e-01 0.00000000e+00]
+- Epoch: 33
 
 ## Model description
 
@@ -474,6 +480,23 @@ The following hyperparameters were used during training:
  0.90790966 0.17633784 0.62937796 0.34917531 0.                nan
  0.         0.4742821  0.2644691  0.         0.94883563 0.91963887
  0.96496596 0.03255799 0.46656807 0.38725811 0.        ]                                                                                                                                                                | 32    |
+| 0.1922     | 0.6796          | 0.3286              | 0.4105                   | 0.8482                      | [0.00000000e+00 7.40810462e-01 8.66118742e-01 4.99364852e-01
+ 5.26878756e-01 3.72552936e-01            nan 4.74841300e-01
+ 5.25079443e-01 1.14740050e-01 8.35714370e-01 6.38207346e-04
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.74621701e-01
+ 0.00000000e+00 0.00000000e+00 7.07495193e-01 1.41447126e-01
+ 4.62213084e-01 3.40642698e-01 0.00000000e+00            nan
+ 0.00000000e+00 3.89053890e-01 2.22898422e-01 0.00000000e+00
+ 8.45321982e-01 7.94221222e-01 9.17872097e-01 1.76532216e-02
+ 2.30290833e-01 2.42034008e-01 0.00000000e+00] | [0.00000000e+00 8.16530181e-01 9.52002646e-01 6.63124748e-01
+ 6.26037619e-01 4.92956588e-01            nan 6.47597208e-01
+ 6.85001856e-01 1.36698809e-01 9.31706346e-01 7.06048482e-04
+ 0.00000000e+00            nan 0.00000000e+00 6.94635901e-01
+ 0.00000000e+00 0.00000000e+00 8.95976958e-01 1.87069749e-01
+ 5.59829035e-01 4.10310044e-01 0.00000000e+00            nan
+ 0.00000000e+00 4.87146698e-01 2.85578447e-01 0.00000000e+00
+ 9.25510256e-01 9.23843473e-01 9.69757916e-01 4.94731348e-02
+ 4.80154090e-01 3.14923397e-01 0.00000000e+00] | 33    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d54e345c03ae28aab14fec6284719783fd338574b243d4bdeef6a35db81e0629
+oid sha256:156179fe75f1817c4eb85cec9f5dd1f797648e504da1b56aa1338a4e600cc8dc
 size 15167588