Training in progress epoch 30

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: 0.2125
-- Validation Loss: 0.6250
-- Validation Mean Iou: 0.3450
-- Validation Mean Accuracy: 0.4185
-- Validation Overall Accuracy: 0.8553
-- Validation Per Category Iou: [0.         0.73159108 0.87022278 0.50404326 0.56244352 0.40464125
-        nan 0.44390964 0.51159512 0.08295183 0.82408288 0.00963956
- 0.                nan 0.         0.58468684 0.         0.
- 0.72766618 0.16108687 0.45500422 0.39801195 0.                nan
- 0.00398936 0.36392878 0.22126126 0.         0.85869236 0.83625848
- 0.92115538 0.01722217 0.19725361 0.34751551 0.        ]
-- Validation Per Category Accuracy: [0.         0.80775127 0.95452298 0.58530378 0.72480765 0.5534954
-        nan 0.62963966 0.69282444 0.09455474 0.90410403 0.01443477
- 0.                nan 0.         0.7591774  0.         0.
- 0.87413938 0.18163073 0.60664188 0.48331242 0.                nan
- 0.00967199 0.44324801 0.27245642 0.         0.9477478  0.91124713
- 0.96830088 0.02370078 0.47974135 0.46998842 0.        ]
-- Epoch: 29
+- Train Loss: 0.1966
+- Validation Loss: 0.6353
+- Validation Mean Iou: 0.3338
+- Validation Mean Accuracy: 0.4147
+- Validation Overall Accuracy: 0.8559
+- Validation Per Category Iou: [0.00000000e+00 7.46226014e-01 8.72110068e-01 4.93922978e-01
+ 5.26412038e-01 4.00773677e-01            nan 4.45688817e-01
+ 5.23812431e-01 1.21566560e-01 8.26532791e-01 1.30318629e-04
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.85128983e-01
+ 0.00000000e+00 0.00000000e+00 7.22504247e-01 1.45035174e-01
+ 4.63070111e-01 3.63454409e-01 0.00000000e+00            nan
+ 1.07188795e-03 3.65266517e-01 2.28463891e-01 0.00000000e+00
+ 8.62155619e-01 8.27775346e-01 9.23668564e-01 2.18692832e-02
+ 2.21420489e-01 3.26921195e-01 0.00000000e+00]
+- Validation Per Category Accuracy: [0.00000000e+00 8.38305417e-01 9.51300548e-01 5.94095086e-01
+ 6.61940847e-01 5.00124050e-01            nan 6.26111992e-01
+ 6.82571110e-01 1.43845718e-01 9.27611840e-01 1.56899663e-04
+ 0.00000000e+00            nan 0.00000000e+00 7.46279093e-01
+ 0.00000000e+00 0.00000000e+00 9.09177452e-01 1.80088323e-01
+ 5.92585802e-01 4.07895465e-01 0.00000000e+00            nan
+ 2.10260723e-03 4.46415580e-01 3.07828843e-01 0.00000000e+00
+ 9.47583864e-01 9.26186116e-01 9.65631281e-01 4.71895084e-02
+ 4.64401183e-01 4.01499481e-01 0.00000000e+00]
+- Epoch: 30
 
 ## Model description
 
@@ -435,6 +441,23 @@ The following hyperparameters were used during training:
  0.87413938 0.18163073 0.60664188 0.48331242 0.                nan
  0.00967199 0.44324801 0.27245642 0.         0.9477478  0.91124713
  0.96830088 0.02370078 0.47974135 0.46998842 0.        ]                                                                                                                                                         | 29    |
+| 0.1966     | 0.6353          | 0.3338              | 0.4147                   | 0.8559                      | [0.00000000e+00 7.46226014e-01 8.72110068e-01 4.93922978e-01
+ 5.26412038e-01 4.00773677e-01            nan 4.45688817e-01
+ 5.23812431e-01 1.21566560e-01 8.26532791e-01 1.30318629e-04
+ 0.00000000e+00 0.00000000e+00 0.00000000e+00 5.85128983e-01
+ 0.00000000e+00 0.00000000e+00 7.22504247e-01 1.45035174e-01
+ 4.63070111e-01 3.63454409e-01 0.00000000e+00            nan
+ 1.07188795e-03 3.65266517e-01 2.28463891e-01 0.00000000e+00
+ 8.62155619e-01 8.27775346e-01 9.23668564e-01 2.18692832e-02
+ 2.21420489e-01 3.26921195e-01 0.00000000e+00] | [0.00000000e+00 8.38305417e-01 9.51300548e-01 5.94095086e-01
+ 6.61940847e-01 5.00124050e-01            nan 6.26111992e-01
+ 6.82571110e-01 1.43845718e-01 9.27611840e-01 1.56899663e-04
+ 0.00000000e+00            nan 0.00000000e+00 7.46279093e-01
+ 0.00000000e+00 0.00000000e+00 9.09177452e-01 1.80088323e-01
+ 5.92585802e-01 4.07895465e-01 0.00000000e+00            nan
+ 2.10260723e-03 4.46415580e-01 3.07828843e-01 0.00000000e+00
+ 9.47583864e-01 9.26186116e-01 9.65631281e-01 4.71895084e-02
+ 4.64401183e-01 4.01499481e-01 0.00000000e+00] | 30    |
 
 
 ### Framework versions

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d3154754c958b11d4b20426148c56e08d725ae6800c7887e6357d1bb64bd73d
+oid sha256:f70274551a057a293e972576f406e9e3f4b33f20393cda2aaae7fb043d7948ff
 size 15167588