Training in progress epoch 28

README.md

@@ -14,30 +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: 0.2340
-- Validation Loss: 0.6300
-- Validation Mean Iou: 0.3327
-- Validation Mean Accuracy: 0.4184
-- Validation Overall Accuracy: 0.8530
-- Validation Per Category Iou: [0.00000000e+00 7.44665882e-01 8.68263319e-01 5.26574132e-01
- 5.88521342e-01 3.51003004e-01            nan 4.33815170e-01
- 5.34720596e-01 1.01138747e-01 8.24056720e-01 2.07010765e-04
- 0.00000000e+00 0.00000000e+00 0.00000000e+00 6.02116780e-01
- 0.00000000e+00 0.00000000e+00 7.13881744e-01 1.36500853e-01
- 4.25515119e-01 4.00566922e-01 0.00000000e+00            nan
- 0.00000000e+00 3.69157017e-01 2.25330495e-01 0.00000000e+00
- 8.55194385e-01 8.27702798e-01 9.19514525e-01 2.51152534e-02
- 1.78530188e-01 3.25860113e-01 0.00000000e+00]
-- Validation Per Category Accuracy: [0.00000000e+00 8.22290631e-01 9.55536279e-01 6.43592714e-01
- 7.08153162e-01 4.50649984e-01            nan 6.10599157e-01
- 7.26377570e-01 1.14861032e-01 9.26899711e-01 2.35349494e-04
- 0.00000000e+00            nan 0.00000000e+00 7.99845170e-01
- 0.00000000e+00 0.00000000e+00 8.29482514e-01 1.51967788e-01
- 6.41329445e-01 4.70916664e-01 0.00000000e+00            nan
- 0.00000000e+00 4.52353087e-01 2.88494453e-01 0.00000000e+00
- 9.49620338e-01 9.19992398e-01 9.70863538e-01 5.53616286e-02
- 4.46687762e-01 4.53320329e-01 0.00000000e+00]
-- Epoch: 27
+- Train Loss: 0.2397
+- Validation Loss: 0.6288
+- Validation Mean Iou: 0.3334
+- Validation Mean Accuracy: 0.4079
+- Validation Overall Accuracy: 0.8479
+- Validation Per Category Iou: [0.00000000e+00 7.50740857e-01 8.57005246e-01 4.57701927e-01
+ 5.76918377e-01 3.94229661e-01            nan 4.28897536e-01
+ 4.95486263e-01 3.27485380e-02 7.89982822e-01 0.00000000e+00
+ 0.00000000e+00            nan 0.00000000e+00 5.88765921e-01
+ 0.00000000e+00 0.00000000e+00 7.15417487e-01 1.49546979e-01
+ 4.39630869e-01 3.01055812e-01 0.00000000e+00            nan
+ 5.49400238e-04 3.73559350e-01 2.04178446e-01 0.00000000e+00
+ 8.58292669e-01 8.28951019e-01 9.12162232e-01 1.81859938e-02
+ 2.05116316e-01 2.88507092e-01 0.00000000e+00]
+- Validation Per Category Accuracy: [0.00000000e+00 8.47527827e-01 9.28350369e-01 5.71594292e-01
+ 7.24101022e-01 6.09780363e-01            nan 5.87257170e-01
+ 6.70956565e-01 3.49404424e-02 9.38259821e-01 0.00000000e+00
+ 0.00000000e+00            nan 0.00000000e+00 7.22055739e-01
+ 0.00000000e+00 0.00000000e+00 8.65879833e-01 1.70168204e-01
+ 6.64439848e-01 3.36883325e-01 0.00000000e+00            nan
+ 8.41042893e-04 4.46411062e-01 2.70110935e-01 0.00000000e+00
+ 9.44186329e-01 8.99719064e-01 9.75261023e-01 2.06668189e-02
+ 4.41253354e-01 3.80569609e-01 0.00000000e+00]
+- Epoch: 28
 
 ## Model description
 
@@ -413,6 +413,23 @@ The following hyperparameters were used during training:
  0.00000000e+00 4.52353087e-01 2.88494453e-01 0.00000000e+00
  9.49620338e-01 9.19992398e-01 9.70863538e-01 5.53616286e-02
  4.46687762e-01 4.53320329e-01 0.00000000e+00] | 27    |
+| 0.2397     | 0.6288          | 0.3334              | 0.4079                   | 0.8479                      | [0.00000000e+00 7.50740857e-01 8.57005246e-01 4.57701927e-01
+ 5.76918377e-01 3.94229661e-01            nan 4.28897536e-01
+ 4.95486263e-01 3.27485380e-02 7.89982822e-01 0.00000000e+00
+ 0.00000000e+00            nan 0.00000000e+00 5.88765921e-01
+ 0.00000000e+00 0.00000000e+00 7.15417487e-01 1.49546979e-01
+ 4.39630869e-01 3.01055812e-01 0.00000000e+00            nan
+ 5.49400238e-04 3.73559350e-01 2.04178446e-01 0.00000000e+00
+ 8.58292669e-01 8.28951019e-01 9.12162232e-01 1.81859938e-02
+ 2.05116316e-01 2.88507092e-01 0.00000000e+00] | [0.00000000e+00 8.47527827e-01 9.28350369e-01 5.71594292e-01
+ 7.24101022e-01 6.09780363e-01            nan 5.87257170e-01
+ 6.70956565e-01 3.49404424e-02 9.38259821e-01 0.00000000e+00
+ 0.00000000e+00            nan 0.00000000e+00 7.22055739e-01
+ 0.00000000e+00 0.00000000e+00 8.65879833e-01 1.70168204e-01
+ 6.64439848e-01 3.36883325e-01 0.00000000e+00            nan
+ 8.41042893e-04 4.46411062e-01 2.70110935e-01 0.00000000e+00
+ 9.44186329e-01 8.99719064e-01 9.75261023e-01 2.06668189e-02
+ 4.41253354e-01 3.80569609e-01 0.00000000e+00] | 28    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:627546c1a1333b6959f6cdca6917122daa66ba1beef4afca5bc84fc8cdb3a41c
+oid sha256:14f6a275fc34bf53a5fb392a830a1469ffafb1f03513a46f8e3666873c4877e0
 size 15167588