Training in progress epoch 35

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: 0.1890
-- Validation Loss: 0.4580
-- Validation Mean Iou: 0.3568
-- Validation Mean Accuracy: 0.4229
-- Validation Overall Accuracy: 0.8760
-- Validation Per Category Iou: [0.         0.81909166 0.87945582 0.84022719 0.85181051 0.44449375
-        nan 0.46587584 0.56531144 0.05849796 0.85834655 0.13038109
- 0.         0.23667513 0.         0.53010343 0.         0.
- 0.70286385 0.08324543 0.31813212 0.48792893 0.                nan
- 0.08722779 0.30648587 0.25211314 0.         0.870058   0.73406627
- 0.93603361 0.09757433 0.04511995 0.17383064 0.        ]
-- Validation Per Category Accuracy: [0.         0.89810904 0.95241393 0.90661944 0.91497772 0.51961956
-        nan 0.65114676 0.83394393 0.06865133 0.95007864 0.14656882
- 0.         0.30775578 0.         0.61461752 0.         0.
- 0.81590436 0.09039982 0.61695222 0.59357779 0.                nan
- 0.09587445 0.39630552 0.2932532  0.         0.94689981 0.84862376
- 0.97537462 0.14406127 0.14217772 0.23057617 0.        ]
-- Epoch: 34
+- Train Loss: 0.1856
+- Validation Loss: 0.4192
+- Validation Mean Iou: 0.3656
+- Validation Mean Accuracy: 0.4327
+- Validation Overall Accuracy: 0.8810
+- Validation Per Category Iou: [0.         0.80138932 0.88262901 0.81089302 0.86535724 0.47775953
+        nan 0.47998869 0.57502219 0.0555217  0.8547197  0.09356223
+ 0.         0.29941446 0.         0.4994726  0.         0.
+ 0.74427779 0.10476547 0.39876117 0.530613   0.                nan
+ 0.06260014 0.28825048 0.26307467 0.         0.86958647 0.76353802
+ 0.93445836 0.20078295 0.02415195 0.18310195 0.        ]
+- Validation Per Category Accuracy: [0.         0.92397697 0.93724033 0.85106091 0.92750395 0.5475243
+        nan 0.64661523 0.82444757 0.07397384 0.94253458 0.12971043
+ 0.         0.39026403 0.         0.56784903 0.         0.
+ 0.88627818 0.11856641 0.54413363 0.67935232 0.                nan
+ 0.0841594  0.35805457 0.30635075 0.         0.93943008 0.852706
+ 0.97181886 0.43921754 0.0738423  0.26218876 0.        ]
+- Epoch: 35
 
 ## Model description
 
@@ -502,6 +502,17 @@ The following hyperparameters were used during training:
  0.81590436 0.09039982 0.61695222 0.59357779 0.                nan
  0.09587445 0.39630552 0.2932532  0.         0.94689981 0.84862376
  0.97537462 0.14406127 0.14217772 0.23057617 0.        ]                                                                                                                                                        | 34    |
+| 0.1856     | 0.4192          | 0.3656              | 0.4327                   | 0.8810                      | [0.         0.80138932 0.88262901 0.81089302 0.86535724 0.47775953
+        nan 0.47998869 0.57502219 0.0555217  0.8547197  0.09356223
+ 0.         0.29941446 0.         0.4994726  0.         0.
+ 0.74427779 0.10476547 0.39876117 0.530613   0.                nan
+ 0.06260014 0.28825048 0.26307467 0.         0.86958647 0.76353802
+ 0.93445836 0.20078295 0.02415195 0.18310195 0.        ]                                                                                                                                                        | [0.         0.92397697 0.93724033 0.85106091 0.92750395 0.5475243
+        nan 0.64661523 0.82444757 0.07397384 0.94253458 0.12971043
+ 0.         0.39026403 0.         0.56784903 0.         0.
+ 0.88627818 0.11856641 0.54413363 0.67935232 0.                nan
+ 0.0841594  0.35805457 0.30635075 0.         0.93943008 0.852706
+ 0.97181886 0.43921754 0.0738423  0.26218876 0.        ]                                                                                                                                                           | 35    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdf9c50383f00fb009d93e3df6c751ffc62ce7d0ba762949ced90734717dfffc
+oid sha256:c43bffb16f2943ec795a6356093032b36188618e89065eaaf20246cf1125aa0b
 size 15167588