Hugging Face's logo

akar49
/
Segformer-MRIseg_model_Mar24

Transformers
TensorFlow
segformer
generated_from_keras_callback
Inference Endpoints
Model card Files Community
akar49 commited on
Commit
4a92739
•
1 Parent(s): 5d3817f

Upload TFSegformerForSemanticSegmentation

Browse files
Files changed (3) hide show
  1. README.md +113 -0
  2. config.json +78 -0
  3. tf_model.h5 +3 -0
README.md ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ base_model: nvidia/segformer-b0-finetuned-ade-512-512
4
+ tags:
5
+ - generated_from_keras_callback
6
+ model-index:
7
+ - name: Segformer-MRIseg_model_Mar24
8
+ results: []
9
+ ---
10
+
11
+ <!-- This model card has been generated automatically according to the information Keras had access to. You should
12
+ probably proofread and complete it, then remove this comment. -->
13
+
14
+ # Segformer-MRIseg_model_Mar24
15
+
16
+ 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.
17
+ It achieves the following results on the evaluation set:
18
+ - Train Loss: 0.0037
19
+ - Validation Loss: 0.0081
20
+ - Epoch: 59
21
+
22
+ ## Model description
23
+
24
+ More information needed
25
+
26
+ ## Intended uses & limitations
27
+
28
+ More information needed
29
+
30
+ ## Training and evaluation data
31
+
32
+ More information needed
33
+
34
+ ## Training procedure
35
+
36
+ ### Training hyperparameters
37
+
38
+ The following hyperparameters were used during training:
39
+ - 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': 0.001, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
40
+ - training_precision: float32
41
+
42
+ ### Training results
43
+
44
+ | Train Loss | Validation Loss | Epoch |
45
+ |:----------:|:---------------:|:-----:|
46
+ | 0.5861 | 0.0929 | 0 |
47
+ | 0.0953 | 0.0606 | 1 |
48
+ | 0.0509 | 0.0350 | 2 |
49
+ | 0.0308 | 0.0231 | 3 |
50
+ | 0.0253 | 0.0219 | 4 |
51
+ | 0.0202 | 0.0179 | 5 |
52
+ | 0.0162 | 0.0184 | 6 |
53
+ | 0.0152 | 0.0188 | 7 |
54
+ | 0.0135 | 0.0157 | 8 |
55
+ | 0.0119 | 0.0170 | 9 |
56
+ | 0.0110 | 0.0150 | 10 |
57
+ | 0.0102 | 0.0157 | 11 |
58
+ | 0.0097 | 0.0137 | 12 |
59
+ | 0.0095 | 0.0141 | 13 |
60
+ | 0.0087 | 0.0118 | 14 |
61
+ | 0.0079 | 0.0116 | 15 |
62
+ | 0.0075 | 0.0119 | 16 |
63
+ | 0.0072 | 0.0109 | 17 |
64
+ | 0.0069 | 0.0118 | 18 |
65
+ | 0.0068 | 0.0104 | 19 |
66
+ | 0.0065 | 0.0108 | 20 |
67
+ | 0.0064 | 0.0124 | 21 |
68
+ | 0.0062 | 0.0095 | 22 |
69
+ | 0.0058 | 0.0111 | 23 |
70
+ | 0.0058 | 0.0094 | 24 |
71
+ | 0.0056 | 0.0111 | 25 |
72
+ | 0.0055 | 0.0125 | 26 |
73
+ | 0.0057 | 0.0104 | 27 |
74
+ | 0.0053 | 0.0096 | 28 |
75
+ | 0.0051 | 0.0105 | 29 |
76
+ | 0.0050 | 0.0103 | 30 |
77
+ | 0.0048 | 0.0091 | 31 |
78
+ | 0.0047 | 0.0097 | 32 |
79
+ | 0.0044 | 0.0094 | 33 |
80
+ | 0.0045 | 0.0092 | 34 |
81
+ | 0.0045 | 0.0093 | 35 |
82
+ | 0.0047 | 0.0088 | 36 |
83
+ | 0.0048 | 0.0089 | 37 |
84
+ | 0.0045 | 0.0108 | 38 |
85
+ | 0.0043 | 0.0088 | 39 |
86
+ | 0.0043 | 0.0090 | 40 |
87
+ | 0.0044 | 0.0106 | 41 |
88
+ | 0.0053 | 0.0100 | 42 |
89
+ | 0.0051 | 0.0102 | 43 |
90
+ | 0.0044 | 0.0097 | 44 |
91
+ | 0.0039 | 0.0088 | 45 |
92
+ | 0.0040 | 0.0097 | 46 |
93
+ | 0.0040 | 0.0089 | 47 |
94
+ | 0.0037 | 0.0095 | 48 |
95
+ | 0.0034 | 0.0085 | 49 |
96
+ | 0.0041 | 0.0082 | 50 |
97
+ | 0.0054 | 0.0098 | 51 |
98
+ | 0.0053 | 0.0085 | 52 |
99
+ | 0.0044 | 0.0086 | 53 |
100
+ | 0.0040 | 0.0082 | 54 |
101
+ | 0.0038 | 0.0082 | 55 |
102
+ | 0.0035 | 0.0092 | 56 |
103
+ | 0.0034 | 0.0090 | 57 |
104
+ | 0.0035 | 0.0079 | 58 |
105
+ | 0.0037 | 0.0081 | 59 |
106
+
107
+
108
+ ### Framework versions
109
+
110
+ - Transformers 4.38.1
111
+ - TensorFlow 2.15.0
112
+ - Datasets 2.18.0
113
+ - Tokenizers 0.15.2
config.json ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "nvidia/segformer-b0-finetuned-ade-512-512",
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": "bg",
32
+ "1": "tumor"
33
+ },
34
+ "image_size": 224,
35
+ "initializer_range": 0.02,
36
+ "label2id": {
37
+ "bg": 0,
38
+ "tumor": 1
39
+ },
40
+ "layer_norm_eps": 1e-06,
41
+ "mlp_ratios": [
42
+ 4,
43
+ 4,
44
+ 4,
45
+ 4
46
+ ],
47
+ "model_type": "segformer",
48
+ "num_attention_heads": [
49
+ 1,
50
+ 2,
51
+ 5,
52
+ 8
53
+ ],
54
+ "num_channels": 3,
55
+ "num_encoder_blocks": 4,
56
+ "patch_sizes": [
57
+ 7,
58
+ 3,
59
+ 3,
60
+ 3
61
+ ],
62
+ "reshape_last_stage": true,
63
+ "semantic_loss_ignore_index": 255,
64
+ "sr_ratios": [
65
+ 8,
66
+ 4,
67
+ 2,
68
+ 1
69
+ ],
70
+ "strides": [
71
+ 4,
72
+ 2,
73
+ 2,
74
+ 2
75
+ ],
76
+ "torch_dtype": "float32",
77
+ "transformers_version": "4.38.1"
78
+ }
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab7210578fe179d676fc1ef2ef7fec35e862e663339182c02b0ba246b6dc7083
3
+ size 15133552