Hugging Face's logo

sayakpaul
/
mit-b0-finetuned-sidewalks

Transformers
TensorFlow
TensorBoard
segformer
generated_from_keras_callback
Inference Endpoints
Model card Files Metrics Community
sayakpaul HF staff commited on
Commit
3089cd5
·
1 Parent(s): efdee5c

Training in progress epoch 5

Browse files
Files changed (2) hide show
  1. README.md +29 -18
  2. tf_model.h5 +1 -1
README.md CHANGED
@@ -14,24 +14,24 @@ probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Train Loss: 0.5810
18
- - Validation Loss: 0.6610
19
- - Validation Mean Iou: 0.2893
20
- - Validation Mean Accuracy: 0.3501
21
- - Validation Overall Accuracy: 0.8173
22
- - Validation Per Category Iou: [0. 0.64601276 0.81866457 0.46535767 0.50543168 0.28373075
23
- nan 0.33004533 0.40404147 0. 0.76223358 0.
24
- 0. nan 0. 0.52641725 0. 0.
25
- 0.65767205 0. 0.39175791 0.25442534 0. nan
26
- 0. 0.30521727 0.03951998 0. 0.82740493 0.75297779
27
- 0.88342457 0. 0.20580056 0.19670152 0. ]
28
- - Validation Per Category Accuracy: [0. 0.73558164 0.95660246 0.55532275 0.61966264 0.32151473
29
- nan 0.47707119 0.54010289 0. 0.91394179 0.
30
- 0. nan 0. 0.63758616 0. 0.
31
- 0.88501875 0. 0.56845175 0.29123233 0. nan
32
- 0. 0.38980592 0.03987322 0. 0.92751577 0.84695264
33
- 0.93293488 0. 0.32582376 0.23722217 0. ]
34
- - Epoch: 4
35
 
36
  ## Model description
37
 
@@ -118,6 +118,17 @@ The following hyperparameters were used during training:
118
  0.88501875 0. 0.56845175 0.29123233 0. nan
119
  0. 0.38980592 0.03987322 0. 0.92751577 0.84695264
120
  0.93293488 0. 0.32582376 0.23722217 0. ] | 4 |
 
 
 
 
 
 
 
 
 
 
 
121
 
122
 
123
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Train Loss: 0.5288
18
+ - Validation Loss: 0.6364
19
+ - Validation Mean Iou: 0.3033
20
+ - Validation Mean Accuracy: 0.3717
21
+ - Validation Overall Accuracy: 0.8260
22
+ - Validation Per Category Iou: [0. 0.64487768 0.8377146 0.48707167 0.50884928 0.34176886
23
+ nan 0.34887555 0.45218372 0. 0.75715898 0.
24
+ 0. nan 0. 0.5222127 0. 0.
25
+ 0.69808156 0. 0.42644563 0.35474225 0. nan
26
+ 0. 0.28867161 0.03742875 0. 0.83332433 0.7818028
27
+ 0.88638015 0.00137015 0.2124537 0.28445749 0. ]
28
+ - Validation Per Category Accuracy: [0. 0.74189035 0.92893266 0.625763 0.62296571 0.54003942
29
+ nan 0.49591369 0.64509343 0. 0.92976992 0.
30
+ 0. nan 0. 0.66209669 0. 0.
31
+ 0.86461114 0. 0.54041026 0.4796133 0. nan
32
+ 0. 0.33899822 0.03746434 0. 0.92987636 0.92582211
33
+ 0.96099073 0.00151698 0.26040449 0.36377671 0. ]
34
+ - Epoch: 5
35
 
36
  ## Model description
37
 
 
118
  0.88501875 0. 0.56845175 0.29123233 0. nan
119
  0. 0.38980592 0.03987322 0. 0.92751577 0.84695264
120
  0.93293488 0. 0.32582376 0.23722217 0. ] | 4 |
121
+ | 0.5288 | 0.6364 | 0.3033 | 0.3717 | 0.8260 | [0. 0.64487768 0.8377146 0.48707167 0.50884928 0.34176886
122
+ nan 0.34887555 0.45218372 0. 0.75715898 0.
123
+ 0. nan 0. 0.5222127 0. 0.
124
+ 0.69808156 0. 0.42644563 0.35474225 0. nan
125
+ 0. 0.28867161 0.03742875 0. 0.83332433 0.7818028
126
+ 0.88638015 0.00137015 0.2124537 0.28445749 0. ] | [0. 0.74189035 0.92893266 0.625763 0.62296571 0.54003942
127
+ nan 0.49591369 0.64509343 0. 0.92976992 0.
128
+ 0. nan 0. 0.66209669 0. 0.
129
+ 0.86461114 0. 0.54041026 0.4796133 0. nan
130
+ 0. 0.33899822 0.03746434 0. 0.92987636 0.92582211
131
+ 0.96099073 0.00151698 0.26040449 0.36377671 0. ] | 5 |
132
 
133
 
134
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:83f7bf17ed571b1d1838402f4cd1ed25d1ead5965e3bbbd289974adff4edf753
3
  size 15167588
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:641d3bcb062b66dadefbdb067180d8f4fd346e3784c56f2b434376479352a078
3
  size 15167588