Hugging Face's logo

smartgmin
/
eyesCare_firstTryEntrnal_mix_model-1

Image Classification
Transformers
TensorFlow
vit
generated_from_keras_callback
Inference Endpoints
Model card Files Community
smartgmin commited on
Commit
568e51a
•
1 Parent(s): 9852f73

Upload TFViTForImageClassification

Browse files
Files changed (3) hide show
  1. README.md +88 -0
  2. config.json +36 -0
  3. tf_model.h5 +3 -0
README.md ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: google/vit-base-patch16-224-in21k
5
+ tags:
6
+ - generated_from_keras_callback
7
+ model-index:
8
+ - name: eyesCare_firstTryEntrnal_mix_model-1
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information Keras had access to. You should
13
+ probably proofread and complete it, then remove this comment. -->
14
+
15
+ # eyesCare_firstTryEntrnal_mix_model-1
16
+
17
+ This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Train Loss: 0.0066
20
+ - Train Accuracy: 0.8616
21
+ - Train Top-3-accuracy: 0.9785
22
+ - Validation Loss: 1.9942
23
+ - Validation Accuracy: 0.8627
24
+ - Validation Top-3-accuracy: 0.9787
25
+ - Epoch: 29
26
+
27
+ ## Model description
28
+
29
+ More information needed
30
+
31
+ ## Intended uses & limitations
32
+
33
+ More information needed
34
+
35
+ ## Training and evaluation data
36
+
37
+ More information needed
38
+
39
+ ## Training procedure
40
+
41
+ ### Training hyperparameters
42
+
43
+ The following hyperparameters were used during training:
44
+ - optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 4e-05, 'decay_steps': 4950, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.001}
45
+ - training_precision: float32
46
+
47
+ ### Training results
48
+
49
+ | Train Loss | Train Accuracy | Train Top-3-accuracy | Validation Loss | Validation Accuracy | Validation Top-3-accuracy | Epoch |
50
+ |:----------:|:--------------:|:--------------------:|:---------------:|:-------------------:|:-------------------------:|:-----:|
51
+ | 1.3981 | 0.3217 | 0.7428 | 1.1812 | 0.4135 | 0.8283 | 0 |
52
+ | 1.1137 | 0.4540 | 0.8600 | 1.0974 | 0.4763 | 0.8802 | 1 |
53
+ | 0.9296 | 0.5034 | 0.8955 | 1.0739 | 0.5231 | 0.9065 | 2 |
54
+ | 0.7444 | 0.5473 | 0.9155 | 1.1126 | 0.5663 | 0.9225 | 3 |
55
+ | 0.5534 | 0.5880 | 0.9285 | 1.1673 | 0.6076 | 0.9342 | 4 |
56
+ | 0.4105 | 0.6261 | 0.9387 | 1.1547 | 0.6422 | 0.9428 | 5 |
57
+ | 0.2830 | 0.6586 | 0.9462 | 1.3119 | 0.6729 | 0.9493 | 6 |
58
+ | 0.1984 | 0.6874 | 0.9519 | 1.3821 | 0.6990 | 0.9540 | 7 |
59
+ | 0.1224 | 0.7104 | 0.9559 | 1.4778 | 0.7213 | 0.9576 | 8 |
60
+ | 0.1021 | 0.7313 | 0.9591 | 1.5426 | 0.7400 | 0.9603 | 9 |
61
+ | 0.1017 | 0.7478 | 0.9615 | 1.6387 | 0.7545 | 0.9625 | 10 |
62
+ | 0.0646 | 0.7613 | 0.9635 | 1.6226 | 0.7678 | 0.9644 | 11 |
63
+ | 0.0500 | 0.7738 | 0.9654 | 1.6646 | 0.7793 | 0.9662 | 12 |
64
+ | 0.0571 | 0.7843 | 0.9669 | 1.7492 | 0.7890 | 0.9675 | 13 |
65
+ | 0.0248 | 0.7935 | 0.9682 | 1.6984 | 0.7978 | 0.9689 | 14 |
66
+ | 0.0185 | 0.8020 | 0.9695 | 1.7302 | 0.8059 | 0.9701 | 15 |
67
+ | 0.0145 | 0.8096 | 0.9707 | 1.7669 | 0.8129 | 0.9712 | 16 |
68
+ | 0.0129 | 0.8163 | 0.9718 | 1.7972 | 0.8193 | 0.9722 | 17 |
69
+ | 0.0116 | 0.8223 | 0.9727 | 1.8276 | 0.8251 | 0.9732 | 18 |
70
+ | 0.0106 | 0.8277 | 0.9736 | 1.8544 | 0.8302 | 0.9739 | 19 |
71
+ | 0.0098 | 0.8326 | 0.9743 | 1.8792 | 0.8348 | 0.9746 | 20 |
72
+ | 0.0091 | 0.8370 | 0.9749 | 1.9012 | 0.8391 | 0.9752 | 21 |
73
+ | 0.0085 | 0.8411 | 0.9755 | 1.9212 | 0.8430 | 0.9758 | 22 |
74
+ | 0.0080 | 0.8448 | 0.9761 | 1.9391 | 0.8465 | 0.9763 | 23 |
75
+ | 0.0076 | 0.8482 | 0.9766 | 1.9547 | 0.8498 | 0.9768 | 24 |
76
+ | 0.0073 | 0.8513 | 0.9770 | 1.9682 | 0.8527 | 0.9772 | 25 |
77
+ | 0.0070 | 0.8542 | 0.9774 | 1.9789 | 0.8555 | 0.9777 | 26 |
78
+ | 0.0068 | 0.8568 | 0.9778 | 1.9871 | 0.8580 | 0.9780 | 27 |
79
+ | 0.0067 | 0.8593 | 0.9782 | 1.9924 | 0.8605 | 0.9784 | 28 |
80
+ | 0.0066 | 0.8616 | 0.9785 | 1.9942 | 0.8627 | 0.9787 | 29 |
81
+
82
+
83
+ ### Framework versions
84
+
85
+ - Transformers 4.44.2
86
+ - TensorFlow 2.15.0
87
+ - Datasets 2.21.0
88
+ - Tokenizers 0.19.1
config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "google/vit-base-patch16-224-in21k",
3
+ "architectures": [
4
+ "ViTForImageClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.0,
7
+ "encoder_stride": 16,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.0,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "Mild",
13
+ "1": "Moderate",
14
+ "2": "No_DR",
15
+ "3": "Proliferative_DR",
16
+ "4": "Severe"
17
+ },
18
+ "image_size": 224,
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 3072,
21
+ "label2id": {
22
+ "Mild": "0",
23
+ "Moderate": "1",
24
+ "No_DR": "2",
25
+ "Proliferative_DR": "3",
26
+ "Severe": "4"
27
+ },
28
+ "layer_norm_eps": 1e-12,
29
+ "model_type": "vit",
30
+ "num_attention_heads": 12,
31
+ "num_channels": 3,
32
+ "num_hidden_layers": 12,
33
+ "patch_size": 16,
34
+ "qkv_bias": true,
35
+ "transformers_version": "4.44.2"
36
+ }
tf_model.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aea5347f52f4fe59913d662f53ac4edb11399e64299df0ccb0a00c795077794b
3
+ size 343478968