Training in progress epoch 39

README.md

@@ -14,24 +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: 0.1824
-- Validation Loss: 0.4277
-- Validation Mean Iou: 0.3516
-- Validation Mean Accuracy: 0.4128
-- Validation Overall Accuracy: 0.8808
-- Validation Per Category Iou: [0.         0.80850849 0.8835188  0.81832156 0.87084804 0.52909381
-        nan 0.48544633 0.57416469 0.06544565 0.86014741 0.09572506
- 0.         0.04364361 0.         0.53546177 0.         0.
- 0.72880369 0.07815572 0.36619794 0.45105441 0.                nan
- 0.02904442 0.31295304 0.268757   0.         0.87009835 0.77016379
- 0.93780115 0.02053909 0.00775031 0.19176263 0.        ]
-- Validation Per Category Accuracy: [0.         0.93462002 0.91830503 0.88981486 0.94415116 0.74919228
-        nan 0.63542345 0.80433651 0.09896256 0.95110348 0.11324871
- 0.         0.04971122 0.         0.61501725 0.         0.
- 0.85270915 0.08732425 0.54361868 0.58993825 0.                nan
- 0.03250764 0.39146001 0.30016676 0.         0.9446708  0.87850239
- 0.97614375 0.02966477 0.01535252 0.27686197 0.        ]
-- Epoch: 38
+- Train Loss: 0.1853
+- Validation Loss: 0.4315
+- Validation Mean Iou: 0.3703
+- Validation Mean Accuracy: 0.4396
+- Validation Overall Accuracy: 0.8843
+- Validation Per Category Iou: [0.         0.82385333 0.88384746 0.82923402 0.87047461 0.44946715
+        nan 0.50195066 0.5526193  0.13775167 0.85419626 0.11663244
+ 0.         0.24123441 0.         0.50296284 0.         0.
+ 0.75525625 0.01028213 0.42676119 0.62702595 0.06111111        nan
+ 0.01903464 0.32208879 0.27514231 0.         0.86733642 0.75225439
+ 0.93747993 0.18324185 0.01936191 0.19849636 0.        ]
+- Validation Per Category Accuracy: [0.         0.90365119 0.95252047 0.86895636 0.95780426 0.51196152
+        nan 0.66021153 0.84474181 0.20306721 0.91545736 0.16898056
+ 0.         0.37891914 0.         0.57321383 0.         0.
+ 0.86912221 0.01029375 0.54801488 0.73707468 0.06340058        nan
+ 0.02149592 0.43258561 0.31736381 0.         0.94644158 0.84006614
+ 0.97630627 0.4108325  0.04714226 0.34773038 0.        ]
+- Epoch: 39
 
 ## Model description
 
@@ -552,6 +552,17 @@ The following hyperparameters were used during training:
  0.85270915 0.08732425 0.54361868 0.58993825 0.                nan
  0.03250764 0.39146001 0.30016676 0.         0.9446708  0.87850239
  0.97614375 0.02966477 0.01535252 0.27686197 0.        ]                                                                                                                                                        | 38    |
+| 0.1853     | 0.4315          | 0.3703              | 0.4396                   | 0.8843                      | [0.         0.82385333 0.88384746 0.82923402 0.87047461 0.44946715
+        nan 0.50195066 0.5526193  0.13775167 0.85419626 0.11663244
+ 0.         0.24123441 0.         0.50296284 0.         0.
+ 0.75525625 0.01028213 0.42676119 0.62702595 0.06111111        nan
+ 0.01903464 0.32208879 0.27514231 0.         0.86733642 0.75225439
+ 0.93747993 0.18324185 0.01936191 0.19849636 0.        ]                                                                                                                                                        | [0.         0.90365119 0.95252047 0.86895636 0.95780426 0.51196152
+        nan 0.66021153 0.84474181 0.20306721 0.91545736 0.16898056
+ 0.         0.37891914 0.         0.57321383 0.         0.
+ 0.86912221 0.01029375 0.54801488 0.73707468 0.06340058        nan
+ 0.02149592 0.43258561 0.31736381 0.         0.94644158 0.84006614
+ 0.97630627 0.4108325  0.04714226 0.34773038 0.        ]                                                                                                                                                        | 39    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dffdb1fc02f63a0b503eaa30aba67b2d24111e5fd2bc5e351ddce90aef902ecd
+oid sha256:12e70a4d1a9f959bebe64484ecd40e0e1f9c2b746bc4a48117ad540dc518735c
 size 15167588