Training in progress epoch 18

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: 0.2734
-- Validation Loss: 0.4359
-- Validation Mean Iou: 0.3324
-- Validation Mean Accuracy: 0.3921
-- Validation Overall Accuracy: 0.8700
-- Validation Per Category Iou: [0.00000000e+00 7.69753150e-01 8.66375446e-01 7.12734804e-01
- 8.60891290e-01 4.75851982e-01            nan 4.34175617e-01
- 5.46685274e-01 3.24503811e-02 8.38429371e-01 6.02288697e-04
- 0.00000000e+00 0.00000000e+00 0.00000000e+00 4.86131744e-01
- 0.00000000e+00 0.00000000e+00 7.27408322e-01 8.04817093e-02
- 3.54733831e-01 5.42941315e-01 0.00000000e+00            nan
- 2.71841412e-04 2.58488818e-01 1.79625596e-01 0.00000000e+00
- 8.66870562e-01 7.48169480e-01 9.27856685e-01 7.49529761e-02
- 5.06564228e-02 1.31117070e-01 0.00000000e+00]
-- Validation Per Category Accuracy: [0.00000000e+00 9.25702577e-01 9.27332221e-01 7.66360378e-01
- 9.04612005e-01 5.93608265e-01            nan 6.05902185e-01
- 7.70458139e-01 3.87911592e-02 9.33542542e-01 1.19000397e-03
- 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.85184718e-01
- 0.00000000e+00 0.00000000e+00 8.77167736e-01 9.19689932e-02
- 5.50758130e-01 6.18987337e-01 0.00000000e+00            nan
- 2.85733125e-04 3.36777474e-01 1.97679266e-01 0.00000000e+00
- 9.34506301e-01 8.56945493e-01 9.73108663e-01 8.09021630e-02
- 2.00250313e-01 1.67017963e-01 0.00000000e+00]
-- Epoch: 17
+- Train Loss: 0.2616
+- Validation Loss: 0.4510
+- Validation Mean Iou: 0.3349
+- Validation Mean Accuracy: 0.4009
+- Validation Overall Accuracy: 0.8726
+- Validation Per Category Iou: [0.         0.79136177 0.88249795 0.78534299 0.86229779 0.47690438
+        nan 0.43122561 0.54187893 0.01320356 0.83487865 0.01648963
+ 0.         0.00144389 0.         0.51996092 0.         0.
+ 0.70356624 0.09487574 0.37358034 0.43221057 0.                nan
+ 0.05030614 0.28552575 0.19602978 0.         0.86214818 0.75003141
+ 0.92402054 0.03367055 0.0295635  0.15961857 0.        ]
+- Validation Per Category Accuracy: [0.         0.9308775  0.94322916 0.85211162 0.90509242 0.57554177
+        nan 0.5785717  0.8193957  0.0157871  0.93896272 0.03232844
+ 0.         0.00144389 0.         0.6718842  0.         0.
+ 0.80146551 0.09782199 0.57800629 0.60986079 0.                nan
+ 0.05345407 0.40116852 0.21407727 0.         0.93948742 0.87904777
+ 0.97272876 0.03561119 0.11522737 0.2649179  0.        ]
+- Epoch: 18
 
 ## Model description
 
@@ -297,6 +291,17 @@ The following hyperparameters were used during training:
  2.85733125e-04 3.36777474e-01 1.97679266e-01 0.00000000e+00
  9.34506301e-01 8.56945493e-01 9.73108663e-01 8.09021630e-02
  2.00250313e-01 1.67017963e-01 0.00000000e+00] | 17    |
+| 0.2616     | 0.4510          | 0.3349              | 0.4009                   | 0.8726                      | [0.         0.79136177 0.88249795 0.78534299 0.86229779 0.47690438
+        nan 0.43122561 0.54187893 0.01320356 0.83487865 0.01648963
+ 0.         0.00144389 0.         0.51996092 0.         0.
+ 0.70356624 0.09487574 0.37358034 0.43221057 0.                nan
+ 0.05030614 0.28552575 0.19602978 0.         0.86214818 0.75003141
+ 0.92402054 0.03367055 0.0295635  0.15961857 0.        ]                                                                                                                                                        | [0.         0.9308775  0.94322916 0.85211162 0.90509242 0.57554177
+        nan 0.5785717  0.8193957  0.0157871  0.93896272 0.03232844
+ 0.         0.00144389 0.         0.6718842  0.         0.
+ 0.80146551 0.09782199 0.57800629 0.60986079 0.                nan
+ 0.05345407 0.40116852 0.21407727 0.         0.93948742 0.87904777
+ 0.97272876 0.03561119 0.11522737 0.2649179  0.        ]                                                                                                                                                        | 18    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:23cdf65f0cc36d1e333b737441c7f6c1ba116f4801583aa4beff14dce865d420
+oid sha256:8d9b70872f985b2ec54d20462c981a6c6caae37234a2e9c01e10af7f5cbc7874
 size 15167588