Training in progress epoch 25

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: 0.2314
-- Validation Loss: 0.4319
-- Validation Mean Iou: 0.3499
-- Validation Mean Accuracy: 0.4182
-- Validation Overall Accuracy: 0.8767
-- Validation Per Category Iou: [0.         0.81268914 0.88090286 0.77985058 0.85570746 0.45828635
-        nan 0.47299524 0.53277529 0.04378213 0.84593985 0.03478787
- 0.         0.1971561  0.         0.54043589 0.         0.
- 0.7359562  0.05923714 0.36875898 0.54834606 0.                nan
- 0.0091682  0.2642792  0.23962131 0.         0.86379601 0.6984222
- 0.9344155  0.19722481 0.0095982  0.16301427 0.        ]
-- Validation Per Category Accuracy: [0.         0.89305484 0.93568687 0.88020793 0.94148981 0.58722529
-        nan 0.61994664 0.81367861 0.04916554 0.95151603 0.06277271
- 0.         0.25453795 0.         0.66359505 0.         0.
- 0.86745813 0.05975395 0.56066626 0.69054391 0.                nan
- 0.01118755 0.32281238 0.27171345 0.         0.9460712  0.8195451
- 0.97224534 0.37103572 0.02778473 0.2253981  0.        ]
-- Epoch: 24
+- Train Loss: 0.2280
+- Validation Loss: 0.4165
+- Validation Mean Iou: 0.3413
+- Validation Mean Accuracy: 0.4040
+- Validation Overall Accuracy: 0.8751
+- Validation Per Category Iou: [0.         0.80627187 0.88428648 0.68558419 0.86189479 0.47772167
+        nan 0.46152926 0.58013249 0.03674413 0.85526933 0.00621328
+ 0.         0.00391914 0.         0.51552043 0.         0.
+ 0.7219038  0.11633219 0.37078391 0.48759114 0.                nan
+ 0.04097689 0.25837403 0.23783935 0.         0.86459819 0.69037029
+ 0.93378787 0.16489662 0.01161391 0.18831714 0.        ]
+- Validation Per Category Accuracy: [0.         0.90036072 0.93838536 0.72247672 0.95316477 0.6161249
+        nan 0.61164372 0.77422776 0.05448805 0.94020362 0.00912336
+ 0.         0.00391914 0.         0.64499705 0.         0.
+ 0.8629953  0.12306992 0.55344734 0.64075339 0.                nan
+ 0.04945381 0.31623508 0.26601584 0.         0.93210162 0.89534679
+ 0.97086444 0.24202332 0.02736754 0.28242179 0.        ]
+- Epoch: 25
 
 ## Model description
 
@@ -386,6 +386,17 @@ The following hyperparameters were used during training:
  0.86745813 0.05975395 0.56066626 0.69054391 0.                nan
  0.01118755 0.32281238 0.27171345 0.         0.9460712  0.8195451
  0.97224534 0.37103572 0.02778473 0.2253981  0.        ]                                                                                                                                                         | 24    |
+| 0.2280     | 0.4165          | 0.3413              | 0.4040                   | 0.8751                      | [0.         0.80627187 0.88428648 0.68558419 0.86189479 0.47772167
+        nan 0.46152926 0.58013249 0.03674413 0.85526933 0.00621328
+ 0.         0.00391914 0.         0.51552043 0.         0.
+ 0.7219038  0.11633219 0.37078391 0.48759114 0.                nan
+ 0.04097689 0.25837403 0.23783935 0.         0.86459819 0.69037029
+ 0.93378787 0.16489662 0.01161391 0.18831714 0.        ]                                                                                                                                                        | [0.         0.90036072 0.93838536 0.72247672 0.95316477 0.6161249
+        nan 0.61164372 0.77422776 0.05448805 0.94020362 0.00912336
+ 0.         0.00391914 0.         0.64499705 0.         0.
+ 0.8629953  0.12306992 0.55344734 0.64075339 0.                nan
+ 0.04945381 0.31623508 0.26601584 0.         0.93210162 0.89534679
+ 0.97086444 0.24202332 0.02736754 0.28242179 0.        ]                                                                                                                                                         | 25    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13d2852299120136436a962a56b4d3090b13322e41b2cb008463706eeb34d28c
+oid sha256:9e92ebf803f184e7fda7779c0a37dbbb4c91c09a7cdced777ade67caa4d44265
 size 15167588