Training in progress epoch 10

README.md

@@ -14,24 +14,30 @@ 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: 0.3970
-- Validation Loss: 0.6094
-- Validation Mean Iou: 0.3154
-- Validation Mean Accuracy: 0.3845
-- Validation Overall Accuracy: 0.8365
-- Validation Per Category Iou: [0.         0.70298046 0.85101569 0.48546331 0.53875077 0.30506197
-        nan 0.36967252 0.46956636 0.         0.78487773 0.
- 0.                nan 0.         0.54510928 0.         0.
- 0.69555357 0.0474544  0.43944078 0.34243937 0.                nan
- 0.         0.31970314 0.07436481 0.         0.84065947 0.79505994
- 0.90045732 0.0134226  0.27125836 0.30161838 0.        ]
-- Validation Per Category Accuracy: [0.         0.78336023 0.94550091 0.59681159 0.65835184 0.43832978
-        nan 0.56126034 0.72465395 0.         0.92553299 0.
- 0.                nan 0.         0.67204825 0.         0.
- 0.84423958 0.04768476 0.62117922 0.41246864 0.                nan
- 0.         0.38035291 0.07866878 0.         0.9420769  0.91782163
- 0.95911561 0.01988386 0.38917934 0.38662485 0.        ]
-- Epoch: 9
+- Train Loss: 0.3827
+- Validation Loss: 0.6001
+- Validation Mean Iou: 0.3153
+- Validation Mean Accuracy: 0.3830
+- Validation Overall Accuracy: 0.8399
+- Validation Per Category Iou: [0.00000000e+00 7.18523940e-01 8.57241195e-01 4.87400335e-01
+ 5.64909227e-01 3.46474032e-01            nan 3.52839201e-01
+ 4.72771964e-01 2.26308345e-04 7.96560839e-01 0.00000000e+00
+ 0.00000000e+00            nan 0.00000000e+00 5.60432300e-01
+ 0.00000000e+00 0.00000000e+00 6.83278176e-01 3.50039777e-02
+ 4.23414561e-01 3.13947966e-01 0.00000000e+00            nan
+ 0.00000000e+00 3.22699644e-01 1.08833077e-01 0.00000000e+00
+ 8.47684853e-01 8.05165122e-01 9.05812333e-01 6.78139068e-03
+ 2.30358320e-01 2.48185158e-01 0.00000000e+00]
+- Validation Per Category Accuracy: [0.00000000e+00 8.00405608e-01 9.43426295e-01 6.48294679e-01
+ 6.81857870e-01 5.18050809e-01            nan 4.57279092e-01
+ 6.27335726e-01 2.26885990e-04 9.00155725e-01 0.00000000e+00
+ 0.00000000e+00            nan 0.00000000e+00 7.09644391e-01
+ 0.00000000e+00 0.00000000e+00 8.82359726e-01 3.50045460e-02
+ 6.04178245e-01 3.33862219e-01 0.00000000e+00            nan
+ 0.00000000e+00 4.02788008e-01 1.21141046e-01 0.00000000e+00
+ 9.46216343e-01 9.21245814e-01 9.55916812e-01 1.34407725e-02
+ 4.56834285e-01 2.96304685e-01 0.00000000e+00]
+- Epoch: 10
 
 ## Model description
 
@@ -173,6 +179,23 @@ The following hyperparameters were used during training:
  0.84423958 0.04768476 0.62117922 0.41246864 0.                nan
  0.         0.38035291 0.07866878 0.         0.9420769  0.91782163
  0.95911561 0.01988386 0.38917934 0.38662485 0.        ]                                                                                                                                                                | 9     |
+| 0.3827     | 0.6001          | 0.3153              | 0.3830                   | 0.8399                      | [0.00000000e+00 7.18523940e-01 8.57241195e-01 4.87400335e-01
+ 5.64909227e-01 3.46474032e-01            nan 3.52839201e-01
+ 4.72771964e-01 2.26308345e-04 7.96560839e-01 0.00000000e+00
+ 0.00000000e+00            nan 0.00000000e+00 5.60432300e-01
+ 0.00000000e+00 0.00000000e+00 6.83278176e-01 3.50039777e-02
+ 4.23414561e-01 3.13947966e-01 0.00000000e+00            nan
+ 0.00000000e+00 3.22699644e-01 1.08833077e-01 0.00000000e+00
+ 8.47684853e-01 8.05165122e-01 9.05812333e-01 6.78139068e-03
+ 2.30358320e-01 2.48185158e-01 0.00000000e+00] | [0.00000000e+00 8.00405608e-01 9.43426295e-01 6.48294679e-01
+ 6.81857870e-01 5.18050809e-01            nan 4.57279092e-01
+ 6.27335726e-01 2.26885990e-04 9.00155725e-01 0.00000000e+00
+ 0.00000000e+00            nan 0.00000000e+00 7.09644391e-01
+ 0.00000000e+00 0.00000000e+00 8.82359726e-01 3.50045460e-02
+ 6.04178245e-01 3.33862219e-01 0.00000000e+00            nan
+ 0.00000000e+00 4.02788008e-01 1.21141046e-01 0.00000000e+00
+ 9.46216343e-01 9.21245814e-01 9.55916812e-01 1.34407725e-02
+ 4.56834285e-01 2.96304685e-01 0.00000000e+00] | 10    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:076d1b80c77c6b2726647076e350ffc51b28c9b607b2001e7ff4ea024006e3e2
+oid sha256:1dae47df6da83fdcea70678255ce9ecea1bc5cbcf3d566e7b039b3183bcecda1
 size 15167588