Training in progress epoch 28

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: 0.2146
-- Validation Loss: 0.4515
-- Validation Mean Iou: 0.3472
-- Validation Mean Accuracy: 0.4129
-- Validation Overall Accuracy: 0.8760
-- Validation Per Category Iou: [0.         0.77768215 0.87939478 0.75655563 0.85872515 0.51912277
-        nan 0.47242163 0.57270343 0.04912978 0.84772409 0.04133148
- 0.         0.28216704 0.         0.50142357 0.         0.
- 0.72646668 0.08708308 0.41888468 0.46264328 0.                nan
- 0.00952928 0.24906863 0.23188316 0.         0.86985544 0.75426408
- 0.93570207 0.         0.00161353 0.15174018 0.        ]
-- Validation Per Category Accuracy: [0.         0.91082524 0.9223558  0.86526048 0.93459861 0.70221741
-        nan 0.61752143 0.76844318 0.064682   0.94790176 0.04432765
- 0.         0.38675743 0.         0.58171337 0.         0.
- 0.86150488 0.08919156 0.64534791 0.76292334 0.                nan
- 0.01094578 0.29978017 0.26035297 0.         0.9390376  0.84793111
- 0.96900737 0.         0.00383813 0.18862775 0.        ]
-- Epoch: 27
+- Train Loss: 0.2245
+- Validation Loss: 0.4819
+- Validation Mean Iou: 0.3481
+- Validation Mean Accuracy: 0.4183
+- Validation Overall Accuracy: 0.8677
+- Validation Per Category Iou: [0.         0.74754716 0.87589221 0.7595096  0.75176585 0.46424109
+        nan 0.43492805 0.55661905 0.04973311 0.85506372 0.1407866
+ 0.         0.15455217 0.         0.4863142  0.         0.
+ 0.72532016 0.16110796 0.37871237 0.54738549 0.                nan
+ 0.03596414 0.27015132 0.27383189 0.         0.87155837 0.74696253
+ 0.93097913 0.03859201 0.03974808 0.19051036 0.        ]
+- Validation Per Category Accuracy: [0.         0.92496748 0.93198417 0.91066332 0.76595448 0.59939476
+        nan 0.57168392 0.76530022 0.06892197 0.94636898 0.19843316
+ 0.         0.30218647 0.         0.5681646  0.         0.
+ 0.83417317 0.18737447 0.68308592 0.66512645 0.                nan
+ 0.03932127 0.32249168 0.30377988 0.         0.94566641 0.86854326
+ 0.97023299 0.04389577 0.11848144 0.26846741 0.        ]
+- Epoch: 28
 
 ## Model description
 
@@ -419,6 +419,17 @@ The following hyperparameters were used during training:
  0.86150488 0.08919156 0.64534791 0.76292334 0.                nan
  0.01094578 0.29978017 0.26035297 0.         0.9390376  0.84793111
  0.96900737 0.         0.00383813 0.18862775 0.        ]                                                                                                                                                        | 27    |
+| 0.2245     | 0.4819          | 0.3481              | 0.4183                   | 0.8677                      | [0.         0.74754716 0.87589221 0.7595096  0.75176585 0.46424109
+        nan 0.43492805 0.55661905 0.04973311 0.85506372 0.1407866
+ 0.         0.15455217 0.         0.4863142  0.         0.
+ 0.72532016 0.16110796 0.37871237 0.54738549 0.                nan
+ 0.03596414 0.27015132 0.27383189 0.         0.87155837 0.74696253
+ 0.93097913 0.03859201 0.03974808 0.19051036 0.        ]                                                                                                                                                         | [0.         0.92496748 0.93198417 0.91066332 0.76595448 0.59939476
+        nan 0.57168392 0.76530022 0.06892197 0.94636898 0.19843316
+ 0.         0.30218647 0.         0.5681646  0.         0.
+ 0.83417317 0.18737447 0.68308592 0.66512645 0.                nan
+ 0.03932127 0.32249168 0.30377988 0.         0.94566641 0.86854326
+ 0.97023299 0.04389577 0.11848144 0.26846741 0.        ]                                                                                                                                                        | 28    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:197db73568e0b0916be2a60f6fbfc90d8a6ef2c2d81fdf01bc26578050285ddf
+oid sha256:a690f664ae42c286f59c0a4144cdf5c8348ff852bd36fcde04c96d72b9eb10bf
 size 15167588