Hugging Face's logo

sayakpaul
/
mit-b0-finetuned-sidewalk-semantic

Image Segmentation
Transformers
TensorFlow
segformer
generated_from_keras_callback
vision
Inference Endpoints
Model card Files Community
1
sayakpaul HF staff commited on
Commit
1a000a0
•
1 Parent(s): e613c3e

add model

Browse files
Files changed (3) hide show
  1. README.md +102 -0
  2. config.json +144 -0
  3. tf_model.h5 +3 -0
README.md ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_keras_callback
5
+ model-index:
6
+ - name: mit-b0-finetuned-sidewalk-semantic
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information Keras had access to. You should
11
+ probably proofread and complete it, then remove this comment. -->
12
+
13
+ # mit-b0-finetuned-sidewalk-semantic
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: 0.2125
18
+ - Validation Loss: 0.5151
19
+ - Epoch: 49
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - optimizer: {'name': 'Adam', 'learning_rate': 6e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
39
+ - training_precision: float32
40
+
41
+ ### Training results
42
+
43
+ | Train Loss | Validation Loss | Epoch |
44
+ |:----------:|:---------------:|:-----:|
45
+ | 2.0785 | 1.1753 | 0 |
46
+ | 1.1312 | 0.8807 | 1 |
47
+ | 0.9315 | 0.7585 | 2 |
48
+ | 0.7952 | 0.7261 | 3 |
49
+ | 0.7273 | 0.6701 | 4 |
50
+ | 0.6603 | 0.6396 | 5 |
51
+ | 0.6198 | 0.6238 | 6 |
52
+ | 0.5958 | 0.5925 | 7 |
53
+ | 0.5378 | 0.5714 | 8 |
54
+ | 0.5236 | 0.5786 | 9 |
55
+ | 0.4960 | 0.5588 | 10 |
56
+ | 0.4633 | 0.5624 | 11 |
57
+ | 0.4562 | 0.5450 | 12 |
58
+ | 0.4167 | 0.5438 | 13 |
59
+ | 0.4100 | 0.5248 | 14 |
60
+ | 0.3947 | 0.5354 | 15 |
61
+ | 0.3867 | 0.5069 | 16 |
62
+ | 0.3803 | 0.5285 | 17 |
63
+ | 0.3696 | 0.5318 | 18 |
64
+ | 0.3386 | 0.5162 | 19 |
65
+ | 0.3349 | 0.5312 | 20 |
66
+ | 0.3233 | 0.5304 | 21 |
67
+ | 0.3328 | 0.5178 | 22 |
68
+ | 0.3140 | 0.5131 | 23 |
69
+ | 0.3081 | 0.5049 | 24 |
70
+ | 0.3046 | 0.5011 | 25 |
71
+ | 0.3209 | 0.5197 | 26 |
72
+ | 0.2966 | 0.5151 | 27 |
73
+ | 0.2829 | 0.5166 | 28 |
74
+ | 0.2968 | 0.5210 | 29 |
75
+ | 0.2818 | 0.5300 | 30 |
76
+ | 0.2739 | 0.5221 | 31 |
77
+ | 0.2602 | 0.5340 | 32 |
78
+ | 0.2570 | 0.5124 | 33 |
79
+ | 0.2557 | 0.5234 | 34 |
80
+ | 0.2593 | 0.5098 | 35 |
81
+ | 0.2582 | 0.5329 | 36 |
82
+ | 0.2439 | 0.5373 | 37 |
83
+ | 0.2413 | 0.5141 | 38 |
84
+ | 0.2423 | 0.5210 | 39 |
85
+ | 0.2340 | 0.5043 | 40 |
86
+ | 0.2244 | 0.5300 | 41 |
87
+ | 0.2246 | 0.4978 | 42 |
88
+ | 0.2270 | 0.5385 | 43 |
89
+ | 0.2254 | 0.5125 | 44 |
90
+ | 0.2176 | 0.5510 | 45 |
91
+ | 0.2194 | 0.5384 | 46 |
92
+ | 0.2136 | 0.5186 | 47 |
93
+ | 0.2121 | 0.5356 | 48 |
94
+ | 0.2125 | 0.5151 | 49 |
95
+
96
+
97
+ ### Framework versions
98
+
99
+ - Transformers 4.21.0.dev0
100
+ - TensorFlow 2.8.0
101
+ - Datasets 2.3.2
102
+ - Tokenizers 0.12.1
config.json ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "nvidia/mit-b0",
3
+ "architectures": [
4
+ "SegformerForSemanticSegmentation"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "classifier_dropout_prob": 0.1,
8
+ "decoder_hidden_size": 256,
9
+ "depths": [
10
+ 2,
11
+ 2,
12
+ 2,
13
+ 2
14
+ ],
15
+ "downsampling_rates": [
16
+ 1,
17
+ 4,
18
+ 8,
19
+ 16
20
+ ],
21
+ "drop_path_rate": 0.1,
22
+ "hidden_act": "gelu",
23
+ "hidden_dropout_prob": 0.0,
24
+ "hidden_sizes": [
25
+ 32,
26
+ 64,
27
+ 160,
28
+ 256
29
+ ],
30
+ "id2label": {
31
+ "0": "unlabeled",
32
+ "1": "flat-road",
33
+ "2": "flat-sidewalk",
34
+ "3": "flat-crosswalk",
35
+ "4": "flat-cyclinglane",
36
+ "5": "flat-parkingdriveway",
37
+ "6": "flat-railtrack",
38
+ "7": "flat-curb",
39
+ "8": "human-person",
40
+ "9": "human-rider",
41
+ "10": "vehicle-car",
42
+ "11": "vehicle-truck",
43
+ "12": "vehicle-bus",
44
+ "13": "vehicle-tramtrain",
45
+ "14": "vehicle-motorcycle",
46
+ "15": "vehicle-bicycle",
47
+ "16": "vehicle-caravan",
48
+ "17": "vehicle-cartrailer",
49
+ "18": "construction-building",
50
+ "19": "construction-door",
51
+ "20": "construction-wall",
52
+ "21": "construction-fenceguardrail",
53
+ "22": "construction-bridge",
54
+ "23": "construction-tunnel",
55
+ "24": "construction-stairs",
56
+ "25": "object-pole",
57
+ "26": "object-trafficsign",
58
+ "27": "object-trafficlight",
59
+ "28": "nature-vegetation",
60
+ "29": "nature-terrain",
61
+ "30": "sky",
62
+ "31": "void-ground",
63
+ "32": "void-dynamic",
64
+ "33": "void-static",
65
+ "34": "void-unclear"
66
+ },
67
+ "image_size": 224,
68
+ "initializer_range": 0.02,
69
+ "label2id": {
70
+ "construction-bridge": 22,
71
+ "construction-building": 18,
72
+ "construction-door": 19,
73
+ "construction-fenceguardrail": 21,
74
+ "construction-stairs": 24,
75
+ "construction-tunnel": 23,
76
+ "construction-wall": 20,
77
+ "flat-crosswalk": 3,
78
+ "flat-curb": 7,
79
+ "flat-cyclinglane": 4,
80
+ "flat-parkingdriveway": 5,
81
+ "flat-railtrack": 6,
82
+ "flat-road": 1,
83
+ "flat-sidewalk": 2,
84
+ "human-person": 8,
85
+ "human-rider": 9,
86
+ "nature-terrain": 29,
87
+ "nature-vegetation": 28,
88
+ "object-pole": 25,
89
+ "object-trafficlight": 27,
90
+ "object-trafficsign": 26,
91
+ "sky": 30,
92
+ "unlabeled": 0,
93
+ "vehicle-bicycle": 15,
94
+ "vehicle-bus": 12,
95
+ "vehicle-car": 10,
96
+ "vehicle-caravan": 16,
97
+ "vehicle-cartrailer": 17,
98
+ "vehicle-motorcycle": 14,
99
+ "vehicle-tramtrain": 13,
100
+ "vehicle-truck": 11,
101
+ "void-dynamic": 32,
102
+ "void-ground": 31,
103
+ "void-static": 33,
104
+ "void-unclear": 34
105
+ },
106
+ "layer_norm_eps": 1e-06,
107
+ "mlp_ratios": [
108
+ 4,
109
+ 4,
110
+ 4,
111
+ 4
112
+ ],
113
+ "model_type": "segformer",
114
+ "num_attention_heads": [
115
+ 1,
116
+ 2,
117
+ 5,
118
+ 8
119
+ ],
120
+ "num_channels": 3,
121
+ "num_encoder_blocks": 4,
122
+ "patch_sizes": [
123
+ 7,
124
+ 3,
125
+ 3,
126
+ 3
127
+ ],
128
+ "reshape_last_stage": true,
129
+ "semantic_loss_ignore_index": 255,
130
+ "sr_ratios": [
131
+ 8,
132
+ 4,
133
+ 2,
134
+ 1
135
+ ],
136
+ "strides": [
137
+ 4,
138
+ 2,
139
+ 2,
140
+ 2
141
+ ],
142
+ "torch_dtype": "float32",
143
+ "transformers_version": "4.21.0.dev0"
144
+ }
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fba28ceaa44bba99481b676c803adfd9ed7e20f186595b34bd9b551f5f3a288a
3
+ size 15167588