Training in progress epoch 26

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: 0.2225
-- Validation Loss: 0.6468
-- Validation Mean Iou: 0.3391
-- Validation Mean Accuracy: 0.4135
-- Validation Overall Accuracy: 0.8494
-- Validation Per Category Iou: [0.         0.70993004 0.86487011 0.52191637 0.57664557 0.3444165
-        nan 0.43372414 0.51182778 0.08290104 0.83020271 0.
- 0.                nan 0.         0.57442713 0.         0.
- 0.73826685 0.15668916 0.47129259 0.34339061 0.                nan
- 0.         0.37774687 0.19761191 0.         0.85313493 0.81669651
- 0.92318268 0.00461872 0.16799491 0.34913299 0.        ]
-- Validation Per Category Accuracy: [0.         0.76188591 0.94953714 0.62857903 0.76788451 0.49450254
-        nan 0.60497833 0.66535259 0.09517867 0.92364467 0.
- 0.                nan 0.         0.74902607 0.         0.
- 0.88590133 0.172701   0.64600271 0.4212137  0.                nan
- 0.         0.45525406 0.24234548 0.         0.93503283 0.93763733
- 0.96721471 0.0103905  0.45387632 0.46408109 0.        ]
-- Epoch: 25
+- Train Loss: 0.2174
+- Validation Loss: 0.6628
+- Validation Mean Iou: 0.3253
+- Validation Mean Accuracy: 0.4085
+- Validation Overall Accuracy: 0.8481
+- Validation Per Category Iou: [0.00000000e+00 7.12004421e-01 8.67849109e-01 5.06086069e-01
+ 5.33272886e-01 3.43273619e-01            nan 4.54065518e-01
+ 5.22321577e-01 1.22861668e-01 8.23942782e-01 4.09584272e-04
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.68874956e-01
+ 0.00000000e+00 0.00000000e+00 7.18493639e-01 1.36007016e-01
+ 4.79722113e-01 3.68024008e-01 0.00000000e+00            nan
+ 0.00000000e+00 3.63955101e-01 1.88599214e-01 0.00000000e+00
+ 8.48893807e-01 8.09523434e-01 9.19749594e-01 6.16287868e-03
+ 1.59932885e-01 2.80999069e-01 0.00000000e+00]
+- Validation Per Category Accuracy: [0.00000000e+00 8.06451570e-01 9.46650239e-01 6.53411928e-01
+ 6.52714160e-01 4.89195197e-01            nan 6.01258726e-01
+ 6.95476161e-01 1.31990925e-01 9.27085524e-01 6.27598651e-04
+ 0.00000000e+00            nan 0.00000000e+00 7.68117571e-01
+ 0.00000000e+00 0.00000000e+00 9.05658822e-01 1.41008573e-01
+ 5.94104088e-01 4.50813623e-01 0.00000000e+00            nan
+ 0.00000000e+00 4.44671381e-01 2.31125198e-01 0.00000000e+00
+ 9.31240381e-01 9.32472058e-01 9.61985230e-01 1.10266532e-02
+ 4.52431726e-01 3.41062626e-01 0.00000000e+00]
+- Epoch: 26
 
 ## Model description
 
@@ -373,6 +379,23 @@ The following hyperparameters were used during training:
  0.88590133 0.172701   0.64600271 0.4212137  0.                nan
  0.         0.45525406 0.24234548 0.         0.93503283 0.93763733
  0.96721471 0.0103905  0.45387632 0.46408109 0.        ]                                                                                                                                                                | 25    |
+| 0.2174     | 0.6628          | 0.3253              | 0.4085                   | 0.8481                      | [0.00000000e+00 7.12004421e-01 8.67849109e-01 5.06086069e-01
+ 5.33272886e-01 3.43273619e-01            nan 4.54065518e-01
+ 5.22321577e-01 1.22861668e-01 8.23942782e-01 4.09584272e-04
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.68874956e-01
+ 0.00000000e+00 0.00000000e+00 7.18493639e-01 1.36007016e-01
+ 4.79722113e-01 3.68024008e-01 0.00000000e+00            nan
+ 0.00000000e+00 3.63955101e-01 1.88599214e-01 0.00000000e+00
+ 8.48893807e-01 8.09523434e-01 9.19749594e-01 6.16287868e-03
+ 1.59932885e-01 2.80999069e-01 0.00000000e+00] | [0.00000000e+00 8.06451570e-01 9.46650239e-01 6.53411928e-01
+ 6.52714160e-01 4.89195197e-01            nan 6.01258726e-01
+ 6.95476161e-01 1.31990925e-01 9.27085524e-01 6.27598651e-04
+ 0.00000000e+00            nan 0.00000000e+00 7.68117571e-01
+ 0.00000000e+00 0.00000000e+00 9.05658822e-01 1.41008573e-01
+ 5.94104088e-01 4.50813623e-01 0.00000000e+00            nan
+ 0.00000000e+00 4.44671381e-01 2.31125198e-01 0.00000000e+00
+ 9.31240381e-01 9.32472058e-01 9.61985230e-01 1.10266532e-02
+ 4.52431726e-01 3.41062626e-01 0.00000000e+00] | 26    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e22d58ac3425a519d6875d85bc02fc730a0c4f8c1b320706ad4cf471a9ee249a
+oid sha256:76bdc44b2068181b6122b730dbac2b9332b7b48923500169a9ca445ef7248e00
 size 15167588