sawthiha commited on
Commit
eed3080
·
1 Parent(s): fb20258

End of training

Browse files
Files changed (3) hide show
  1. README.md +19 -73
  2. model.safetensors +3 -0
  3. training_args.bin +3 -0
README.md CHANGED
@@ -2,30 +2,27 @@
2
  license: other
3
  base_model: nvidia/segformer-b0-finetuned-ade-512-512
4
  tags:
5
- - generated_from_keras_callback
6
  - vision
7
  - image-segmentation
 
8
  model-index:
9
  - name: segformer-b0-finetuned-deprem-satellite
10
  results: []
11
- pipeline_tag: image-segmentation
12
- widget:
13
- - src: https://datasets-server.huggingface.co/assets/deprem-ml/deprem_satellite_semantic_whu_dataset/--/default/train/3/image/image.jpg
14
- example_title: Example 1
15
- - src: https://datasets-server.huggingface.co/assets/deprem-ml/deprem_satellite_semantic_whu_dataset/--/default/train/9/image/image.jpg
16
- example_title: Example 2
17
  ---
18
 
19
- <!-- This model card has been generated automatically according to the information Keras had access to. You should
20
- probably proofread and complete it, then remove this comment. -->
21
 
22
  # segformer-b0-finetuned-deprem-satellite
23
 
24
- This model is a fine-tuned version of [nvidia/segformer-b0-finetuned-ade-512-512](https://huggingface.co/nvidia/segformer-b0-finetuned-ade-512-512) on an unknown dataset.
25
  It achieves the following results on the evaluation set:
26
- - Train Loss: 0.0405
27
- - Validation Loss: 0.0344
28
- - Epoch: 49
 
 
 
29
 
30
  ## Model description
31
 
@@ -44,68 +41,17 @@ More information needed
44
  ### Training hyperparameters
45
 
46
  The following hyperparameters were used during training:
47
- - optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': 5e-05, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
48
- - training_precision: float32
49
-
50
- ### Training results
51
-
52
- | Train Loss | Validation Loss | Epoch |
53
- |:----------:|:---------------:|:-----:|
54
- | 0.1619 | 0.0524 | 0 |
55
- | 0.0943 | 0.0437 | 1 |
56
- | 0.0811 | 0.0402 | 2 |
57
- | 0.0756 | 0.0390 | 3 |
58
- | 0.0714 | 0.0365 | 4 |
59
- | 0.0676 | 0.0367 | 5 |
60
- | 0.0648 | 0.0361 | 6 |
61
- | 0.0627 | 0.0352 | 7 |
62
- | 0.0617 | 0.0423 | 8 |
63
- | 0.0597 | 0.0348 | 9 |
64
- | 0.0582 | 0.0338 | 10 |
65
- | 0.0575 | 0.0340 | 11 |
66
- | 0.0553 | 0.0338 | 12 |
67
- | 0.0551 | 0.0328 | 13 |
68
- | 0.0542 | 0.0353 | 14 |
69
- | 0.0541 | 0.0348 | 15 |
70
- | 0.0526 | 0.0325 | 16 |
71
- | 0.0512 | 0.0326 | 17 |
72
- | 0.0511 | 0.0338 | 18 |
73
- | 0.0506 | 0.0338 | 19 |
74
- | 0.0501 | 0.0326 | 20 |
75
- | 0.0488 | 0.0339 | 21 |
76
- | 0.0488 | 0.0329 | 22 |
77
- | 0.0481 | 0.0335 | 23 |
78
- | 0.0475 | 0.0327 | 24 |
79
- | 0.0469 | 0.0335 | 25 |
80
- | 0.0463 | 0.0331 | 26 |
81
- | 0.0478 | 0.0326 | 27 |
82
- | 0.0455 | 0.0331 | 28 |
83
- | 0.0511 | 0.0328 | 29 |
84
- | 0.0454 | 0.0327 | 30 |
85
- | 0.0447 | 0.0330 | 31 |
86
- | 0.0466 | 0.0341 | 32 |
87
- | 0.0445 | 0.0331 | 33 |
88
- | 0.0441 | 0.0333 | 34 |
89
- | 0.0438 | 0.0337 | 35 |
90
- | 0.0441 | 0.0347 | 36 |
91
- | 0.0463 | 0.0334 | 37 |
92
- | 0.0446 | 0.0336 | 38 |
93
- | 0.0430 | 0.0337 | 39 |
94
- | 0.0425 | 0.0342 | 40 |
95
- | 0.0422 | 0.0339 | 41 |
96
- | 0.0426 | 0.0348 | 42 |
97
- | 0.0418 | 0.0339 | 43 |
98
- | 0.0416 | 0.0339 | 44 |
99
- | 0.0414 | 0.0343 | 45 |
100
- | 0.0423 | 0.0337 | 46 |
101
- | 0.0408 | 0.0353 | 47 |
102
- | 0.0405 | 0.0343 | 48 |
103
- | 0.0405 | 0.0344 | 49 |
104
-
105
 
106
  ### Framework versions
107
 
108
  - Transformers 4.36.2
109
- - TensorFlow 2.15.0
110
  - Datasets 2.16.1
111
- - Tokenizers 0.15.0
 
2
  license: other
3
  base_model: nvidia/segformer-b0-finetuned-ade-512-512
4
  tags:
 
5
  - vision
6
  - image-segmentation
7
+ - generated_from_trainer
8
  model-index:
9
  - name: segformer-b0-finetuned-deprem-satellite
10
  results: []
 
 
 
 
 
 
11
  ---
12
 
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # segformer-b0-finetuned-deprem-satellite
17
 
18
+ This model is a fine-tuned version of [nvidia/segformer-b0-finetuned-ade-512-512](https://huggingface.co/nvidia/segformer-b0-finetuned-ade-512-512) on the deprem-ml/deprem_satellite_semantic_whu_dataset dataset.
19
  It achieves the following results on the evaluation set:
20
+ - eval_loss: 0.0849
21
+ - eval_runtime: 64.056
22
+ - eval_samples_per_second: 16.173
23
+ - eval_steps_per_second: 4.043
24
+ - epoch: 4.18
25
+ - step: 3960
26
 
27
  ## Model description
28
 
 
41
  ### Training hyperparameters
42
 
43
  The following hyperparameters were used during training:
44
+ - learning_rate: 5e-05
45
+ - train_batch_size: 5
46
+ - eval_batch_size: 4
47
+ - seed: 42
48
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
49
+ - lr_scheduler_type: linear
50
+ - num_epochs: 50
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
51
 
52
  ### Framework versions
53
 
54
  - Transformers 4.36.2
55
+ - Pytorch 2.1.2
56
  - Datasets 2.16.1
57
+ - Tokenizers 0.15.0
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bfe3c78af33b55e3899e0632e124223ab0602f9fa7c9298bcacfcd653fa38275
3
+ size 14884776
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:014646daa30947a2a5e78b514576ccb3f04faed51db92120b962d4c9ba2c81aa
3
+ size 4728