Training in progress epoch 41

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: 0.1750
-- Validation Loss: 0.4506
-- Validation Mean Iou: 0.3588
-- Validation Mean Accuracy: 0.4247
-- Validation Overall Accuracy: 0.8777
-- Validation Per Category Iou: [0.         0.8085732  0.87850952 0.80281118 0.80588458 0.43532471
-        nan 0.49235495 0.56953178 0.04144004 0.86063777 0.1253508
- 0.         0.29125623 0.         0.56867524 0.         0.
- 0.72796854 0.03398534 0.34825502 0.58349517 0.04278075        nan
- 0.02356575 0.3145996  0.25091045 0.         0.86821676 0.7616297
- 0.94167999 0.0760155  0.01097801 0.17674597 0.        ]
-- Validation Per Category Accuracy: [0.         0.8587409  0.96059527 0.84914689 0.95691624 0.49065318
-        nan 0.64206647 0.80599476 0.04330176 0.94759276 0.13288378
- 0.         0.49401815 0.         0.68717916 0.         0.
- 0.83929688 0.03427065 0.62310627 0.69531488 0.04610951        nan
- 0.02716663 0.42729112 0.28005142 0.         0.94607512 0.87723316
- 0.97355068 0.11214495 0.02653317 0.23750462 0.        ]
-- Epoch: 40
+- Train Loss: 0.1856
+- Validation Loss: 0.4630
+- Validation Mean Iou: 0.3468
+- Validation Mean Accuracy: 0.4169
+- Validation Overall Accuracy: 0.8702
+- Validation Per Category Iou: [0.         0.79680775 0.87408906 0.80508886 0.7836208  0.54111501
+        nan 0.46906575 0.51571781 0.05906675 0.84646062 0.12752339
+ 0.         0.02326951 0.         0.49396166 0.         0.
+ 0.70301274 0.10455404 0.36326525 0.45432265 0.00288184        nan
+ 0.06241517 0.29796314 0.27257073 0.         0.85824781 0.70264792
+ 0.93738763 0.10944047 0.07468448 0.16378606 0.        ]
+- Validation Per Category Accuracy: [0.         0.92066362 0.93733642 0.8457505  0.80518061 0.72000372
+        nan 0.61879522 0.83036713 0.090212   0.93724404 0.17978977
+ 0.         0.03259076 0.         0.52749727 0.         0.
+ 0.80827055 0.12335237 0.6834038  0.55220964 0.00288184        nan
+ 0.08086247 0.38878944 0.30794886 0.         0.93395437 0.8773043
+ 0.97470902 0.14777538 0.20984564 0.21967583 0.        ]
+- Epoch: 41
 
 ## Model description
 
@@ -574,6 +574,17 @@ The following hyperparameters were used during training:
  0.83929688 0.03427065 0.62310627 0.69531488 0.04610951        nan
  0.02716663 0.42729112 0.28005142 0.         0.94607512 0.87723316
  0.97355068 0.11214495 0.02653317 0.23750462 0.        ]                                                                                                                                                        | 40    |
+| 0.1856     | 0.4630          | 0.3468              | 0.4169                   | 0.8702                      | [0.         0.79680775 0.87408906 0.80508886 0.7836208  0.54111501
+        nan 0.46906575 0.51571781 0.05906675 0.84646062 0.12752339
+ 0.         0.02326951 0.         0.49396166 0.         0.
+ 0.70301274 0.10455404 0.36326525 0.45432265 0.00288184        nan
+ 0.06241517 0.29796314 0.27257073 0.         0.85824781 0.70264792
+ 0.93738763 0.10944047 0.07468448 0.16378606 0.        ]                                                                                                                                                        | [0.         0.92066362 0.93733642 0.8457505  0.80518061 0.72000372
+        nan 0.61879522 0.83036713 0.090212   0.93724404 0.17978977
+ 0.         0.03259076 0.         0.52749727 0.         0.
+ 0.80827055 0.12335237 0.6834038  0.55220964 0.00288184        nan
+ 0.08086247 0.38878944 0.30794886 0.         0.93395437 0.8773043
+ 0.97470902 0.14777538 0.20984564 0.21967583 0.        ]                                                                                                                                                         | 41    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1ace38a79d81954f8c5397e82a777919145a4b3c07091baef322d32f0eedfc1b
+oid sha256:9ea6eb4d5571158b6e8b8e3cff15148862107e45ee792cac4e6351c3c3beaeca
 size 15167588