Hugging Face's logo

nateraw
/
mit-b0-finetuned-sidewalks-v2

Transformers
TensorFlow
segformer
generated_from_keras_callback
Inference Endpoints
Model card Files Community
nateraw commited on
Commit
340931e
•
1 Parent(s): af2a09e

Training in progress epoch 13

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: 0.3460
18
- - Validation Loss: 0.4342
19
- - Validation Mean Iou: 0.3234
20
- - Validation Mean Accuracy: 0.3852
21
- - Validation Overall Accuracy: 0.8669
22
- - Validation Per Category Iou: [0. 0.76828673 0.86958873 0.66044471 0.84588115 0.46323947
23
- nan 0.41208499 0.54202812 0. 0.82543751 0.
24
- 0. 0. 0. 0.50071248 0. 0.
25
- 0.72333932 0.0173886 0.36535728 0.5284402 0. nan
26
- 0. 0.24239821 0.13456635 0. 0.86084123 0.73217705
27
- 0.92386442 0.09545854 0.04193608 0.11945951 0. ]
28
- - Validation Per Category Accuracy: [0. 0.92666259 0.91906703 0.74134089 0.92518489 0.60022437
29
- nan 0.56316038 0.77045814 0. 0.93600314 0.
30
- 0. 0. 0. 0.61358664 0. 0.
31
- 0.87835072 0.01757469 0.57608316 0.64108174 0. nan
32
- 0. 0.30432247 0.13750695 0. 0.93332326 0.85806371
33
- 0.96442783 0.10753599 0.15152274 0.14552189 0. ]
34
- - Epoch: 12
35
 
36
  ## Model description
37
 
@@ -218,6 +218,17 @@ The following hyperparameters were used during training:
218
  0.87835072 0.01757469 0.57608316 0.64108174 0. nan
219
  0. 0.30432247 0.13750695 0. 0.93332326 0.85806371
220
  0.96442783 0.10753599 0.15152274 0.14552189 0. ] | 12 |
 
 
 
 
 
 
 
 
 
 
 
221
 
222
 
223
  ### 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: 0.3146
18
+ - Validation Loss: 0.4175
19
+ - Validation Mean Iou: 0.3339
20
+ - Validation Mean Accuracy: 0.3995
21
+ - Validation Overall Accuracy: 0.8745
22
+ - Validation Per Category Iou: [0. 0.81054591 0.88286867 0.68551149 0.86089895 0.4562385
23
+ nan 0.4522713 0.55496016 0.01456189 0.83576109 0.
24
+ 0. 0. 0. 0.50709788 0. 0.
25
+ 0.73464008 0.00175153 0.35021502 0.57263292 0. nan
26
+ 0. 0.25185222 0.14419755 0. 0.85952374 0.70281003
27
+ 0.9270307 0.17660456 0.04867831 0.18762581 0. ]
28
+ - Validation Per Category Accuracy: [0. 0.9092016 0.94168672 0.86545289 0.89611216 0.55273728
29
+ nan 0.61409823 0.76682349 0.01569689 0.92776282 0.
30
+ 0. 0. 0. 0.59972229 0. 0.
31
+ 0.86700656 0.00175747 0.54181633 0.67419762 0. nan
32
+ 0. 0.3252672 0.14789466 0. 0.9316378 0.88743565
33
+ 0.97060047 0.33277846 0.15319149 0.25967892 0. ]
34
+ - Epoch: 13
35
 
36
  ## Model description
37
 
 
218
  0.87835072 0.01757469 0.57608316 0.64108174 0. nan
219
  0. 0.30432247 0.13750695 0. 0.93332326 0.85806371
220
  0.96442783 0.10753599 0.15152274 0.14552189 0. ] | 12 |
221
+ | 0.3146 | 0.4175 | 0.3339 | 0.3995 | 0.8745 | [0. 0.81054591 0.88286867 0.68551149 0.86089895 0.4562385
222
+ nan 0.4522713 0.55496016 0.01456189 0.83576109 0.
223
+ 0. 0. 0. 0.50709788 0. 0.
224
+ 0.73464008 0.00175153 0.35021502 0.57263292 0. nan
225
+ 0. 0.25185222 0.14419755 0. 0.85952374 0.70281003
226
+ 0.9270307 0.17660456 0.04867831 0.18762581 0. ] | [0. 0.9092016 0.94168672 0.86545289 0.89611216 0.55273728
227
+ nan 0.61409823 0.76682349 0.01569689 0.92776282 0.
228
+ 0. 0. 0. 0.59972229 0. 0.
229
+ 0.86700656 0.00175747 0.54181633 0.67419762 0. nan
230
+ 0. 0.3252672 0.14789466 0. 0.9316378 0.88743565
231
+ 0.97060047 0.33277846 0.15319149 0.25967892 0. ] | 13 |
232
 
233
 
234
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6d51fc1bc9331a5c1205c0ddd9b3b065322fc1c7ab6186dcd5911d57b362b028
3
  size 15167588
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3caaecd9ad82aba4cca0c46592605db39abe1893254d6be66380e1e659c22e3
3
  size 15167588