Training in progress epoch 32

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: 0.1993
-- Validation Loss: 0.4191
-- Validation Mean Iou: 0.3525
-- Validation Mean Accuracy: 0.4026
-- Validation Overall Accuracy: 0.8855
-- Validation Per Category Iou: [0.00000000e+00 8.12143438e-01 8.82519501e-01 8.32421151e-01
- 8.74313051e-01 4.81253823e-01            nan 4.87073361e-01
- 5.86132068e-01 9.62771937e-02 8.59982957e-01 8.85474149e-02
- 0.00000000e+00 2.01491034e-04 0.00000000e+00 5.35389616e-01
- 0.00000000e+00 0.00000000e+00 7.53505814e-01 3.69389833e-02
- 3.98315791e-01 5.86352445e-01 0.00000000e+00            nan
- 3.86967641e-02 2.99523304e-01 2.23544639e-01 0.00000000e+00
- 8.66545952e-01 7.59345221e-01 9.36605085e-01 5.82816319e-04
- 7.04036476e-04 1.95231882e-01 0.00000000e+00]
-- Validation Per Category Accuracy: [0.00000000e+00 9.18027876e-01 9.52593301e-01 8.60861913e-01
- 9.17933036e-01 5.59645609e-01            nan 6.69381444e-01
- 7.81217462e-01 1.31348669e-01 9.42924301e-01 1.03431178e-01
- 0.00000000e+00 2.06270627e-04 0.00000000e+00 6.30669865e-01
- 0.00000000e+00 0.00000000e+00 8.97879175e-01 3.70010043e-02
- 4.89103277e-01 6.59469339e-01 0.00000000e+00            nan
- 4.58931358e-02 3.79932245e-01 2.46004725e-01 0.00000000e+00
- 9.44760882e-01 8.53162772e-01 9.75178979e-01 5.96566854e-04
- 1.75219024e-03 2.87445529e-01 0.00000000e+00]
-- Epoch: 31
+- Train Loss: 0.2068
+- Validation Loss: 0.4805
+- Validation Mean Iou: 0.3370
+- Validation Mean Accuracy: 0.3952
+- Validation Overall Accuracy: 0.8643
+- Validation Per Category Iou: [0.         0.77056757 0.8601312  0.79546358 0.80826542 0.46090981
+        nan 0.47734482 0.58905088 0.03181978 0.85901467 0.01694625
+ 0.         0.00549451 0.         0.48326241 0.         0.
+ 0.71413255 0.08548594 0.355285   0.56037404 0.                nan
+ 0.11377479 0.28155688 0.23155416 0.         0.84077004 0.62872483
+ 0.94074387 0.04323906 0.00477968 0.16213294 0.        ]
+- Validation Per Category Accuracy: [0.         0.86625295 0.93033124 0.83741848 0.95175277 0.58905634
+        nan 0.60932022 0.77904824 0.04077582 0.93578138 0.01963507
+ 0.         0.00556931 0.         0.57342422 0.         0.
+ 0.8469629  0.09113733 0.63002638 0.69225687 0.                nan
+ 0.12851397 0.34756471 0.25621873 0.         0.94994706 0.68278681
+ 0.96967504 0.05913709 0.01677096 0.23138435 0.        ]
+- Epoch: 32
 
 ## Model description
 
@@ -475,6 +469,17 @@ The following hyperparameters were used during training:
  4.58931358e-02 3.79932245e-01 2.46004725e-01 0.00000000e+00
  9.44760882e-01 8.53162772e-01 9.75178979e-01 5.96566854e-04
  1.75219024e-03 2.87445529e-01 0.00000000e+00] | 31    |
+| 0.2068     | 0.4805          | 0.3370              | 0.3952                   | 0.8643                      | [0.         0.77056757 0.8601312  0.79546358 0.80826542 0.46090981
+        nan 0.47734482 0.58905088 0.03181978 0.85901467 0.01694625
+ 0.         0.00549451 0.         0.48326241 0.         0.
+ 0.71413255 0.08548594 0.355285   0.56037404 0.                nan
+ 0.11377479 0.28155688 0.23155416 0.         0.84077004 0.62872483
+ 0.94074387 0.04323906 0.00477968 0.16213294 0.        ]                                                                                                                                                        | [0.         0.86625295 0.93033124 0.83741848 0.95175277 0.58905634
+        nan 0.60932022 0.77904824 0.04077582 0.93578138 0.01963507
+ 0.         0.00556931 0.         0.57342422 0.         0.
+ 0.8469629  0.09113733 0.63002638 0.69225687 0.                nan
+ 0.12851397 0.34756471 0.25621873 0.         0.94994706 0.68278681
+ 0.96967504 0.05913709 0.01677096 0.23138435 0.        ]                                                                                                                                                        | 32    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4fd415857999ae5c5a4c45f457166fa85509c7e66f00f7296fb89873586b6086
+oid sha256:73ddaa1a0dd0ee7f5b76c0d923cdad45cf43c556ccf4024601d2740d9c51e94e
 size 15167588