Hugging Face's logo

greathero
/
mit-b0-finetuned-sidewalks

Transformers
TensorFlow
segformer
generated_from_keras_callback
Model card Files
xet
 Community
greathero commited on
Commit
12f7965
·
1 Parent(s): 2e54842

Training in progress epoch 49

Browse files
Files changed (2) hide show
  1. README.md +55 -54
  2. tf_model.h5 +1 -1
README.md CHANGED
@@ -15,82 +15,82 @@ probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
- - Train Loss: 0.2038
19
- - Validation Loss: 0.5407
20
- - Validation Mean Iou: 0.3318
21
- - Validation Mean Accuracy: 0.3995
22
- - Validation Overall Accuracy: 0.8701
23
  - Validation Accuracy Unlabeled: 0.0
24
- - Validation Accuracy Flat-road: 0.9396
25
- - Validation Accuracy Flat-sidewalk: 0.9574
26
- - Validation Accuracy Flat-crosswalk: 0.6559
27
- - Validation Accuracy Flat-cyclinglane: 0.7289
28
- - Validation Accuracy Flat-parkingdriveway: 0.6166
29
  - Validation Accuracy Flat-railtrack: nan
30
- - Validation Accuracy Flat-curb: 0.5573
31
- - Validation Accuracy Human-person: 0.6550
32
- - Validation Accuracy Human-rider: 0.1861
33
- - Validation Accuracy Vehicle-car: 0.9233
34
- - Validation Accuracy Vehicle-truck: 0.0852
35
  - Validation Accuracy Vehicle-bus: 0.0
36
  - Validation Accuracy Vehicle-tramtrain: nan
37
  - Validation Accuracy Vehicle-motorcycle: 0.0
38
- - Validation Accuracy Vehicle-bicycle: 0.6396
39
  - Validation Accuracy Vehicle-caravan: 0.0
40
  - Validation Accuracy Vehicle-cartrailer: 0.0
41
- - Validation Accuracy Construction-building: 0.9241
42
- - Validation Accuracy Construction-door: 0.0504
43
- - Validation Accuracy Construction-wall: 0.4811
44
- - Validation Accuracy Construction-fenceguardrail: 0.4240
45
  - Validation Accuracy Construction-bridge: 0.0
46
  - Validation Accuracy Construction-tunnel: nan
47
- - Validation Accuracy Construction-stairs: 0.0001
48
- - Validation Accuracy Object-pole: 0.3993
49
- - Validation Accuracy Object-trafficsign: 0.4905
50
  - Validation Accuracy Object-trafficlight: 0.0
51
- - Validation Accuracy Nature-vegetation: 0.9225
52
- - Validation Accuracy Nature-terrain: 0.9142
53
- - Validation Accuracy Sky: 0.9672
54
- - Validation Accuracy Void-ground: 0.0155
55
- - Validation Accuracy Void-dynamic: 0.0175
56
- - Validation Accuracy Void-static: 0.2326
57
  - Validation Accuracy Void-unclear: 0.0
58
  - Validation Iou Unlabeled: 0.0
59
- - Validation Iou Flat-road: 0.8068
60
- - Validation Iou Flat-sidewalk: 0.9021
61
- - Validation Iou Flat-crosswalk: 0.3095
62
- - Validation Iou Flat-cyclinglane: 0.7049
63
- - Validation Iou Flat-parkingdriveway: 0.4679
64
  - Validation Iou Flat-railtrack: nan
65
- - Validation Iou Flat-curb: 0.4504
66
- - Validation Iou Human-person: 0.4687
67
- - Validation Iou Human-rider: 0.1686
68
- - Validation Iou Vehicle-car: 0.8264
69
- - Validation Iou Vehicle-truck: 0.0847
70
  - Validation Iou Vehicle-bus: 0.0
71
  - Validation Iou Vehicle-tramtrain: nan
72
  - Validation Iou Vehicle-motorcycle: 0.0
73
- - Validation Iou Vehicle-bicycle: 0.4721
74
  - Validation Iou Vehicle-caravan: 0.0
75
  - Validation Iou Vehicle-cartrailer: 0.0
76
- - Validation Iou Construction-building: 0.6820
77
- - Validation Iou Construction-door: 0.0482
78
- - Validation Iou Construction-wall: 0.3826
79
- - Validation Iou Construction-fenceguardrail: 0.3686
80
  - Validation Iou Construction-bridge: 0.0
81
  - Validation Iou Construction-tunnel: nan
82
- - Validation Iou Construction-stairs: 0.0001
83
- - Validation Iou Object-pole: 0.3217
84
- - Validation Iou Object-trafficsign: 0.3838
85
  - Validation Iou Object-trafficlight: 0.0
86
- - Validation Iou Nature-vegetation: 0.8636
87
- - Validation Iou Nature-terrain: 0.7679
88
- - Validation Iou Sky: 0.9297
89
- - Validation Iou Void-ground: 0.0114
90
- - Validation Iou Void-dynamic: 0.0156
91
- - Validation Iou Void-static: 0.1809
92
  - Validation Iou Void-unclear: 0.0
93
- - Epoch: 48
94
 
95
  ## Model description
96
 
@@ -165,6 +165,7 @@ The following hyperparameters were used during training:
165
  | 0.2021 | 0.5369 | 0.3409 | 0.4049 | 0.8753 | 0.0 | 0.9249 | 0.9608 | 0.5473 | 0.7979 | 0.6534 | nan | 0.5548 | 0.6555 | 0.1623 | 0.9311 | 0.1317 | 0.0 | nan | 0.0 | 0.6821 | 0.0 | 0.0 | 0.9151 | 0.0621 | 0.5285 | 0.4378 | 0.0 | nan | 0.0029 | 0.4205 | 0.5321 | 0.0 | 0.9301 | 0.8972 | 0.9677 | 0.0025 | 0.0281 | 0.2294 | 0.0 | 0.0 | 0.8155 | 0.8981 | 0.3979 | 0.7396 | 0.4890 | nan | 0.4634 | 0.4928 | 0.1352 | 0.8283 | 0.1299 | 0.0 | nan | 0.0 | 0.4784 | 0.0 | 0.0 | 0.6949 | 0.0602 | 0.4199 | 0.3775 | 0.0 | nan | 0.0018 | 0.3280 | 0.3792 | 0.0 | 0.8698 | 0.7691 | 0.9283 | 0.0020 | 0.0237 | 0.1856 | 0.0 | 46 |
166
  | 0.2056 | 0.5118 | 0.3421 | 0.4052 | 0.8747 | 0.0 | 0.9441 | 0.9624 | 0.5219 | 0.7370 | 0.6816 | nan | 0.5360 | 0.6510 | 0.2372 | 0.9316 | 0.0854 | 0.0 | nan | 0.0 | 0.6607 | 0.0 | 0.0 | 0.8952 | 0.0777 | 0.5614 | 0.4327 | 0.0 | nan | 0.0 | 0.4403 | 0.4597 | 0.0 | 0.9350 | 0.8918 | 0.9650 | 0.0312 | 0.0296 | 0.2984 | 0.0 | 0.0 | 0.8039 | 0.9033 | 0.4243 | 0.7185 | 0.4949 | nan | 0.4494 | 0.4876 | 0.1929 | 0.8314 | 0.0849 | 0.0 | nan | 0.0 | 0.4694 | 0.0 | 0.0 | 0.7160 | 0.0697 | 0.3959 | 0.3751 | 0.0 | nan | 0.0 | 0.3319 | 0.3668 | 0.0 | 0.8638 | 0.7612 | 0.9299 | 0.0246 | 0.0251 | 0.2269 | 0.0 | 47 |
167
  | 0.2038 | 0.5407 | 0.3318 | 0.3995 | 0.8701 | 0.0 | 0.9396 | 0.9574 | 0.6559 | 0.7289 | 0.6166 | nan | 0.5573 | 0.6550 | 0.1861 | 0.9233 | 0.0852 | 0.0 | nan | 0.0 | 0.6396 | 0.0 | 0.0 | 0.9241 | 0.0504 | 0.4811 | 0.4240 | 0.0 | nan | 0.0001 | 0.3993 | 0.4905 | 0.0 | 0.9225 | 0.9142 | 0.9672 | 0.0155 | 0.0175 | 0.2326 | 0.0 | 0.0 | 0.8068 | 0.9021 | 0.3095 | 0.7049 | 0.4679 | nan | 0.4504 | 0.4687 | 0.1686 | 0.8264 | 0.0847 | 0.0 | nan | 0.0 | 0.4721 | 0.0 | 0.0 | 0.6820 | 0.0482 | 0.3826 | 0.3686 | 0.0 | nan | 0.0001 | 0.3217 | 0.3838 | 0.0 | 0.8636 | 0.7679 | 0.9297 | 0.0114 | 0.0156 | 0.1809 | 0.0 | 48 |
 
168
 
169
 
170
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Train Loss: 0.1964
19
+ - Validation Loss: 0.5617
20
+ - Validation Mean Iou: 0.3482
21
+ - Validation Mean Accuracy: 0.4212
22
+ - Validation Overall Accuracy: 0.8714
23
  - Validation Accuracy Unlabeled: 0.0
24
+ - Validation Accuracy Flat-road: 0.9191
25
+ - Validation Accuracy Flat-sidewalk: 0.9583
26
+ - Validation Accuracy Flat-crosswalk: 0.5878
27
+ - Validation Accuracy Flat-cyclinglane: 0.7730
28
+ - Validation Accuracy Flat-parkingdriveway: 0.6079
29
  - Validation Accuracy Flat-railtrack: nan
30
+ - Validation Accuracy Flat-curb: 0.5577
31
+ - Validation Accuracy Human-person: 0.6959
32
+ - Validation Accuracy Human-rider: 0.2221
33
+ - Validation Accuracy Vehicle-car: 0.9300
34
+ - Validation Accuracy Vehicle-truck: 0.1310
35
  - Validation Accuracy Vehicle-bus: 0.0
36
  - Validation Accuracy Vehicle-tramtrain: nan
37
  - Validation Accuracy Vehicle-motorcycle: 0.0
38
+ - Validation Accuracy Vehicle-bicycle: 0.6927
39
  - Validation Accuracy Vehicle-caravan: 0.0
40
  - Validation Accuracy Vehicle-cartrailer: 0.0
41
+ - Validation Accuracy Construction-building: 0.9088
42
+ - Validation Accuracy Construction-door: 0.0611
43
+ - Validation Accuracy Construction-wall: 0.4825
44
+ - Validation Accuracy Construction-fenceguardrail: 0.4370
45
  - Validation Accuracy Construction-bridge: 0.0
46
  - Validation Accuracy Construction-tunnel: nan
47
+ - Validation Accuracy Construction-stairs: 0.0
48
+ - Validation Accuracy Object-pole: 0.4339
49
+ - Validation Accuracy Object-trafficsign: 0.4711
50
  - Validation Accuracy Object-trafficlight: 0.0
51
+ - Validation Accuracy Nature-vegetation: 0.9172
52
+ - Validation Accuracy Nature-terrain: 0.9151
53
+ - Validation Accuracy Sky: 0.9692
54
+ - Validation Accuracy Void-ground: 0.5121
55
+ - Validation Accuracy Void-dynamic: 0.0265
56
+ - Validation Accuracy Void-static: 0.2676
57
  - Validation Accuracy Void-unclear: 0.0
58
  - Validation Iou Unlabeled: 0.0
59
+ - Validation Iou Flat-road: 0.7902
60
+ - Validation Iou Flat-sidewalk: 0.9044
61
+ - Validation Iou Flat-crosswalk: 0.4201
62
+ - Validation Iou Flat-cyclinglane: 0.6963
63
+ - Validation Iou Flat-parkingdriveway: 0.4879
64
  - Validation Iou Flat-railtrack: nan
65
+ - Validation Iou Flat-curb: 0.4607
66
+ - Validation Iou Human-person: 0.4926
67
+ - Validation Iou Human-rider: 0.1864
68
+ - Validation Iou Vehicle-car: 0.8243
69
+ - Validation Iou Vehicle-truck: 0.1278
70
  - Validation Iou Vehicle-bus: 0.0
71
  - Validation Iou Vehicle-tramtrain: nan
72
  - Validation Iou Vehicle-motorcycle: 0.0
73
+ - Validation Iou Vehicle-bicycle: 0.4759
74
  - Validation Iou Vehicle-caravan: 0.0
75
  - Validation Iou Vehicle-cartrailer: 0.0
76
+ - Validation Iou Construction-building: 0.6973
77
+ - Validation Iou Construction-door: 0.0588
78
+ - Validation Iou Construction-wall: 0.3754
79
+ - Validation Iou Construction-fenceguardrail: 0.3760
80
  - Validation Iou Construction-bridge: 0.0
81
  - Validation Iou Construction-tunnel: nan
82
+ - Validation Iou Construction-stairs: 0.0
83
+ - Validation Iou Object-pole: 0.3330
84
+ - Validation Iou Object-trafficsign: 0.3555
85
  - Validation Iou Object-trafficlight: 0.0
86
+ - Validation Iou Nature-vegetation: 0.8600
87
+ - Validation Iou Nature-terrain: 0.7533
88
+ - Validation Iou Sky: 0.9288
89
+ - Validation Iou Void-ground: 0.3089
90
+ - Validation Iou Void-dynamic: 0.0223
91
+ - Validation Iou Void-static: 0.2055
92
  - Validation Iou Void-unclear: 0.0
93
+ - Epoch: 49
94
 
95
  ## Model description
96
 
 
165
  | 0.2021 | 0.5369 | 0.3409 | 0.4049 | 0.8753 | 0.0 | 0.9249 | 0.9608 | 0.5473 | 0.7979 | 0.6534 | nan | 0.5548 | 0.6555 | 0.1623 | 0.9311 | 0.1317 | 0.0 | nan | 0.0 | 0.6821 | 0.0 | 0.0 | 0.9151 | 0.0621 | 0.5285 | 0.4378 | 0.0 | nan | 0.0029 | 0.4205 | 0.5321 | 0.0 | 0.9301 | 0.8972 | 0.9677 | 0.0025 | 0.0281 | 0.2294 | 0.0 | 0.0 | 0.8155 | 0.8981 | 0.3979 | 0.7396 | 0.4890 | nan | 0.4634 | 0.4928 | 0.1352 | 0.8283 | 0.1299 | 0.0 | nan | 0.0 | 0.4784 | 0.0 | 0.0 | 0.6949 | 0.0602 | 0.4199 | 0.3775 | 0.0 | nan | 0.0018 | 0.3280 | 0.3792 | 0.0 | 0.8698 | 0.7691 | 0.9283 | 0.0020 | 0.0237 | 0.1856 | 0.0 | 46 |
166
  | 0.2056 | 0.5118 | 0.3421 | 0.4052 | 0.8747 | 0.0 | 0.9441 | 0.9624 | 0.5219 | 0.7370 | 0.6816 | nan | 0.5360 | 0.6510 | 0.2372 | 0.9316 | 0.0854 | 0.0 | nan | 0.0 | 0.6607 | 0.0 | 0.0 | 0.8952 | 0.0777 | 0.5614 | 0.4327 | 0.0 | nan | 0.0 | 0.4403 | 0.4597 | 0.0 | 0.9350 | 0.8918 | 0.9650 | 0.0312 | 0.0296 | 0.2984 | 0.0 | 0.0 | 0.8039 | 0.9033 | 0.4243 | 0.7185 | 0.4949 | nan | 0.4494 | 0.4876 | 0.1929 | 0.8314 | 0.0849 | 0.0 | nan | 0.0 | 0.4694 | 0.0 | 0.0 | 0.7160 | 0.0697 | 0.3959 | 0.3751 | 0.0 | nan | 0.0 | 0.3319 | 0.3668 | 0.0 | 0.8638 | 0.7612 | 0.9299 | 0.0246 | 0.0251 | 0.2269 | 0.0 | 47 |
167
  | 0.2038 | 0.5407 | 0.3318 | 0.3995 | 0.8701 | 0.0 | 0.9396 | 0.9574 | 0.6559 | 0.7289 | 0.6166 | nan | 0.5573 | 0.6550 | 0.1861 | 0.9233 | 0.0852 | 0.0 | nan | 0.0 | 0.6396 | 0.0 | 0.0 | 0.9241 | 0.0504 | 0.4811 | 0.4240 | 0.0 | nan | 0.0001 | 0.3993 | 0.4905 | 0.0 | 0.9225 | 0.9142 | 0.9672 | 0.0155 | 0.0175 | 0.2326 | 0.0 | 0.0 | 0.8068 | 0.9021 | 0.3095 | 0.7049 | 0.4679 | nan | 0.4504 | 0.4687 | 0.1686 | 0.8264 | 0.0847 | 0.0 | nan | 0.0 | 0.4721 | 0.0 | 0.0 | 0.6820 | 0.0482 | 0.3826 | 0.3686 | 0.0 | nan | 0.0001 | 0.3217 | 0.3838 | 0.0 | 0.8636 | 0.7679 | 0.9297 | 0.0114 | 0.0156 | 0.1809 | 0.0 | 48 |
168
+ | 0.1964 | 0.5617 | 0.3482 | 0.4212 | 0.8714 | 0.0 | 0.9191 | 0.9583 | 0.5878 | 0.7730 | 0.6079 | nan | 0.5577 | 0.6959 | 0.2221 | 0.9300 | 0.1310 | 0.0 | nan | 0.0 | 0.6927 | 0.0 | 0.0 | 0.9088 | 0.0611 | 0.4825 | 0.4370 | 0.0 | nan | 0.0 | 0.4339 | 0.4711 | 0.0 | 0.9172 | 0.9151 | 0.9692 | 0.5121 | 0.0265 | 0.2676 | 0.0 | 0.0 | 0.7902 | 0.9044 | 0.4201 | 0.6963 | 0.4879 | nan | 0.4607 | 0.4926 | 0.1864 | 0.8243 | 0.1278 | 0.0 | nan | 0.0 | 0.4759 | 0.0 | 0.0 | 0.6973 | 0.0588 | 0.3754 | 0.3760 | 0.0 | nan | 0.0 | 0.3330 | 0.3555 | 0.0 | 0.8600 | 0.7533 | 0.9288 | 0.3089 | 0.0223 | 0.2055 | 0.0 | 49 |
169
 
170
 
171
  ### Framework versions
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6e292ed6f0bf5c041967875891ce203c3223ffa629b6c039e84b53722c07bc6b
3
  size 15167476
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2108fcf41d0a298a307521b0569b5b68850ba96ccf753c69a9eafffb52dd94a
3
  size 15167476