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
098b7bf
·
1 Parent(s): 0b2de8f

Training in progress epoch 8

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: 0.4528
18
- - Validation Loss: 0.6340
19
- - Validation Mean Iou: 0.2980
20
- - Validation Mean Accuracy: 0.3664
21
- - Validation Overall Accuracy: 0.8212
22
- - Validation Per Category Iou: [0. 0.60868123 0.83498291 0.18287132 0.46939835 0.31058578
23
- nan 0.34162709 0.445366 0. 0.78966215 0.
24
- 0. nan 0. 0.53583212 0. 0.
25
- 0.71233622 0.03447214 0.47235409 0.37419598 0. nan
26
- 0. 0.32268508 0.05312127 0. 0.83874416 0.79217023
27
- 0.89975806 0.00192312 0.20492869 0.31166384 0. ]
28
- - Validation Per Category Accuracy: [0. 0.70588336 0.94249106 0.19298309 0.73275474 0.44094168
29
- nan 0.48341533 0.71859918 0. 0.90854187 0.
30
- 0. nan 0. 0.6752697 0. 0.
31
- 0.88312382 0.03451747 0.65575793 0.42127597 0. nan
32
- 0. 0.39094462 0.05356577 0. 0.95291962 0.86480438
33
- 0.95973926 0.00251199 0.29803261 0.40525743 0. ]
34
- - Epoch: 7
35
 
36
  ## Model description
37
 
@@ -151,6 +151,17 @@ The following hyperparameters were used during training:
151
  0.88312382 0.03451747 0.65575793 0.42127597 0. nan
152
  0. 0.39094462 0.05356577 0. 0.95291962 0.86480438
153
  0.95973926 0.00251199 0.29803261 0.40525743 0. ] | 7 |
 
 
 
 
 
 
 
 
 
 
 
154
 
155
 
156
  ### 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: 0.4458
18
+ - Validation Loss: 0.6356
19
+ - Validation Mean Iou: 0.3066
20
+ - Validation Mean Accuracy: 0.3759
21
+ - Validation Overall Accuracy: 0.8311
22
+ - Validation Per Category Iou: [0. 0.66186021 0.83364406 0.48364478 0.53627121 0.27582606
23
+ nan 0.35739504 0.41076225 0. 0.77850446 0.
24
+ 0. nan 0. 0.50299945 0. 0.
25
+ 0.70340595 0.01741996 0.40137463 0.32885851 0. nan
26
+ 0. 0.32114603 0.05439069 0. 0.84655176 0.80081688
27
+ 0.90314194 0.00292704 0.24599153 0.34524384 0. ]
28
+ - Validation Per Category Accuracy: [0. 0.74699777 0.95226432 0.65030589 0.70566846 0.35272168
29
+ nan 0.47019568 0.71372001 0. 0.91011138 0.
30
+ 0. nan 0. 0.62189092 0. 0.
31
+ 0.87945472 0.01742109 0.52016264 0.36958738 0. nan
32
+ 0. 0.3993222 0.05508716 0. 0.928178 0.89821483
33
+ 0.96116851 0.0033765 0.40579212 0.46633448 0. ]
34
+ - Epoch: 8
35
 
36
  ## Model description
37
 
 
151
  0.88312382 0.03451747 0.65575793 0.42127597 0. nan
152
  0. 0.39094462 0.05356577 0. 0.95291962 0.86480438
153
  0.95973926 0.00251199 0.29803261 0.40525743 0. ] | 7 |
154
+ | 0.4458 | 0.6356 | 0.3066 | 0.3759 | 0.8311 | [0. 0.66186021 0.83364406 0.48364478 0.53627121 0.27582606
155
+ nan 0.35739504 0.41076225 0. 0.77850446 0.
156
+ 0. nan 0. 0.50299945 0. 0.
157
+ 0.70340595 0.01741996 0.40137463 0.32885851 0. nan
158
+ 0. 0.32114603 0.05439069 0. 0.84655176 0.80081688
159
+ 0.90314194 0.00292704 0.24599153 0.34524384 0. ] | [0. 0.74699777 0.95226432 0.65030589 0.70566846 0.35272168
160
+ nan 0.47019568 0.71372001 0. 0.91011138 0.
161
+ 0. nan 0. 0.62189092 0. 0.
162
+ 0.87945472 0.01742109 0.52016264 0.36958738 0. nan
163
+ 0. 0.3993222 0.05508716 0. 0.928178 0.89821483
164
+ 0.96116851 0.0033765 0.40579212 0.46633448 0. ] | 8 |
165
 
166
 
167
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7f8646cb606f2ec9def37bd221e0af59c073e1487adfef0e399ff0d6fd5f067f
3
  size 15167588
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7296d22da245469f541f0a1edbea989595cec88e7b88389d543dfb8272b18a39
3
  size 15167588