whyoke commited on
Commit
93769c1
·
verified ·
1 Parent(s): 41b14bc

whyoke/segmentation_model_50ep

Browse files
Files changed (1) hide show
  1. README.md +103 -0
README.md ADDED
@@ -0,0 +1,103 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: other
4
+ base_model: nvidia/mit-b0
5
+ tags:
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: segmentation_model_50ep
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # segmentation_model_50ep
16
+
17
+ This model is a fine-tuned version of [nvidia/mit-b0](https://huggingface.co/nvidia/mit-b0) on the None dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.0063
20
+ - Mean Iou: 0.9981
21
+ - Mean Accuracy: 1.0
22
+ - Overall Accuracy: 1.0
23
+ - Per Category Iou: [0.9980539089681099]
24
+ - Per Category Accuracy: [1.0]
25
+
26
+ ## Model description
27
+
28
+ More information needed
29
+
30
+ ## Intended uses & limitations
31
+
32
+ More information needed
33
+
34
+ ## Training and evaluation data
35
+
36
+ More information needed
37
+
38
+ ## Training procedure
39
+
40
+ ### Training hyperparameters
41
+
42
+ The following hyperparameters were used during training:
43
+ - learning_rate: 6e-05
44
+ - train_batch_size: 16
45
+ - eval_batch_size: 16
46
+ - seed: 42
47
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
+ - lr_scheduler_type: linear
49
+ - num_epochs: 50
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss | Mean Iou | Mean Accuracy | Overall Accuracy | Per Category Iou | Per Category Accuracy |
54
+ |:-------------:|:-------:|:----:|:---------------:|:--------:|:-------------:|:----------------:|:--------------------:|:---------------------:|
55
+ | 0.049 | 1.2195 | 100 | 0.0429 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
56
+ | 0.029 | 2.4390 | 200 | 0.0274 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
57
+ | 0.0171 | 3.6585 | 300 | 0.0192 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
58
+ | 0.0158 | 4.8780 | 400 | 0.0187 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
59
+ | 0.019 | 6.0976 | 500 | 0.0169 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
60
+ | 0.013 | 7.3171 | 600 | 0.0125 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
61
+ | 0.0131 | 8.5366 | 700 | 0.0124 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
62
+ | 0.0111 | 9.7561 | 800 | 0.0101 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
63
+ | 0.0089 | 10.9756 | 900 | 0.0102 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
64
+ | 0.0106 | 12.1951 | 1000 | 0.0088 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
65
+ | 0.0093 | 13.4146 | 1100 | 0.0084 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
66
+ | 0.0088 | 14.6341 | 1200 | 0.0079 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
67
+ | 0.0084 | 15.8537 | 1300 | 0.0080 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
68
+ | 0.0089 | 17.0732 | 1400 | 0.0077 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
69
+ | 0.0087 | 18.2927 | 1500 | 0.0069 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
70
+ | 0.0072 | 19.5122 | 1600 | 0.0075 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
71
+ | 0.0087 | 20.7317 | 1700 | 0.0068 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
72
+ | 0.0094 | 21.9512 | 1800 | 0.0070 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
73
+ | 0.0074 | 23.1707 | 1900 | 0.0070 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
74
+ | 0.0075 | 24.3902 | 2000 | 0.0069 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
75
+ | 0.007 | 25.6098 | 2100 | 0.0064 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
76
+ | 0.0053 | 26.8293 | 2200 | 0.0065 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
77
+ | 0.0072 | 28.0488 | 2300 | 0.0063 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
78
+ | 0.0082 | 29.2683 | 2400 | 0.0065 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
79
+ | 0.0065 | 30.4878 | 2500 | 0.0066 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
80
+ | 0.0054 | 31.7073 | 2600 | 0.0065 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
81
+ | 0.0079 | 32.9268 | 2700 | 0.0066 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
82
+ | 0.006 | 34.1463 | 2800 | 0.0064 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
83
+ | 0.0053 | 35.3659 | 2900 | 0.0063 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
84
+ | 0.0059 | 36.5854 | 3000 | 0.0064 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
85
+ | 0.0061 | 37.8049 | 3100 | 0.0066 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
86
+ | 0.007 | 39.0244 | 3200 | 0.0064 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
87
+ | 0.0058 | 40.2439 | 3300 | 0.0063 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
88
+ | 0.0055 | 41.4634 | 3400 | 0.0062 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
89
+ | 0.0068 | 42.6829 | 3500 | 0.0064 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
90
+ | 0.0058 | 43.9024 | 3600 | 0.0063 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
91
+ | 0.0061 | 45.1220 | 3700 | 0.0064 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
92
+ | 0.003 | 46.3415 | 3800 | 0.0063 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
93
+ | 0.0058 | 47.5610 | 3900 | 0.0063 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
94
+ | 0.0087 | 48.7805 | 4000 | 0.0063 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
95
+ | 0.006 | 50.0 | 4100 | 0.0063 | 0.9981 | 1.0 | 1.0 | [0.9980539089681099] | [1.0] |
96
+
97
+
98
+ ### Framework versions
99
+
100
+ - Transformers 4.46.3
101
+ - Pytorch 2.2.0
102
+ - Datasets 2.4.0
103
+ - Tokenizers 0.20.3