nomsgadded commited on
Commit
437fb4d
1 Parent(s): d7fe4a1

Model save

Browse files
Files changed (2) hide show
  1. README.md +2 -79
  2. model.safetensors +1 -1
README.md CHANGED
@@ -2,8 +2,6 @@
2
  license: other
3
  base_model: nvidia/mit-b0
4
  tags:
5
- - image-segmentation
6
- - vision
7
  - generated_from_trainer
8
  model-index:
9
  - name: Segments
@@ -15,82 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # Segments
17
 
18
- This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the segments/sidewalk-semantic dataset.
19
- It achieves the following results on the evaluation set:
20
- - Loss: 1.0916
21
- - Mean Iou: 0.1657
22
- - Mean Accuracy: 0.2139
23
- - Overall Accuracy: 0.7523
24
- - Accuracy Unlabeled: nan
25
- - Accuracy Flat-road: 0.7670
26
- - Accuracy Flat-sidewalk: 0.9044
27
- - Accuracy Flat-crosswalk: 0.0
28
- - Accuracy Flat-cyclinglane: 0.7817
29
- - Accuracy Flat-parkingdriveway: 0.0531
30
- - Accuracy Flat-railtrack: nan
31
- - Accuracy Flat-curb: 0.0008
32
- - Accuracy Human-person: 0.0
33
- - Accuracy Human-rider: 0.0
34
- - Accuracy Vehicle-car: 0.8784
35
- - Accuracy Vehicle-truck: 0.0
36
- - Accuracy Vehicle-bus: 0.0
37
- - Accuracy Vehicle-tramtrain: 0.0
38
- - Accuracy Vehicle-motorcycle: 0.0
39
- - Accuracy Vehicle-bicycle: 0.0
40
- - Accuracy Vehicle-caravan: 0.0
41
- - Accuracy Vehicle-cartrailer: 0.0
42
- - Accuracy Construction-building: 0.8946
43
- - Accuracy Construction-door: 0.0
44
- - Accuracy Construction-wall: 0.0018
45
- - Accuracy Construction-fenceguardrail: 0.0
46
- - Accuracy Construction-bridge: 0.0
47
- - Accuracy Construction-tunnel: nan
48
- - Accuracy Construction-stairs: 0.0
49
- - Accuracy Object-pole: 0.0
50
- - Accuracy Object-trafficsign: 0.0
51
- - Accuracy Object-trafficlight: 0.0
52
- - Accuracy Nature-vegetation: 0.9132
53
- - Accuracy Nature-terrain: 0.8337
54
- - Accuracy Sky: 0.8152
55
- - Accuracy Void-ground: 0.0
56
- - Accuracy Void-dynamic: 0.0
57
- - Accuracy Void-static: 0.0
58
- - Accuracy Void-unclear: 0.0
59
- - Iou Unlabeled: nan
60
- - Iou Flat-road: 0.5357
61
- - Iou Flat-sidewalk: 0.7678
62
- - Iou Flat-crosswalk: 0.0
63
- - Iou Flat-cyclinglane: 0.6199
64
- - Iou Flat-parkingdriveway: 0.0504
65
- - Iou Flat-railtrack: nan
66
- - Iou Flat-curb: 0.0008
67
- - Iou Human-person: 0.0
68
- - Iou Human-rider: 0.0
69
- - Iou Vehicle-car: 0.6256
70
- - Iou Vehicle-truck: 0.0
71
- - Iou Vehicle-bus: 0.0
72
- - Iou Vehicle-tramtrain: 0.0
73
- - Iou Vehicle-motorcycle: 0.0
74
- - Iou Vehicle-bicycle: 0.0
75
- - Iou Vehicle-caravan: 0.0
76
- - Iou Vehicle-cartrailer: 0.0
77
- - Iou Construction-building: 0.5403
78
- - Iou Construction-door: 0.0
79
- - Iou Construction-wall: 0.0018
80
- - Iou Construction-fenceguardrail: 0.0
81
- - Iou Construction-bridge: 0.0
82
- - Iou Construction-tunnel: nan
83
- - Iou Construction-stairs: 0.0
84
- - Iou Object-pole: 0.0
85
- - Iou Object-trafficsign: 0.0
86
- - Iou Object-trafficlight: 0.0
87
- - Iou Nature-vegetation: 0.7516
88
- - Iou Nature-terrain: 0.6290
89
- - Iou Sky: 0.7785
90
- - Iou Void-ground: 0.0
91
- - Iou Void-dynamic: 0.0
92
- - Iou Void-static: 0.0
93
- - Iou Void-unclear: 0.0
94
 
95
  ## Model description
96
 
@@ -117,7 +40,7 @@ The following hyperparameters were used during training:
117
  - total_train_batch_size: 16
118
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
119
  - lr_scheduler_type: linear
120
- - num_epochs: 15.0
121
 
122
  ### Training results
123
 
 
2
  license: other
3
  base_model: nvidia/mit-b0
4
  tags:
 
 
5
  - generated_from_trainer
6
  model-index:
7
  - name: Segments
 
13
 
14
  # Segments
15
 
16
+ This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on an unknown dataset.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
17
 
18
  ## Model description
19
 
 
40
  - total_train_batch_size: 16
41
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
  - lr_scheduler_type: linear
43
+ - num_epochs: 20.0
44
 
45
  ### Training results
46
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:07c325f0633951c20606c8587feb9a5c6a589f9dc340c017c3d629cc292dc446
3
  size 14918708
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1991f97e9f07cefc8de8ccafb51a37a0bb6099018615724a0da76dc677245a5
3
  size 14918708