Sejin0503 commited on
Commit
f8b9e88
1 Parent(s): cbcc347
README.md ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: google/vit-base-patch16-224-in21k
4
+ tags:
5
+ - image-classification
6
+ - ViT
7
+ - generated_from_trainer
8
+ metrics:
9
+ - accuracy
10
+ model-index:
11
+ - name: vit-base-beans-demo-v5
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # vit-base-beans-demo-v5
19
+
20
+ This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the beans dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 0.1959
23
+ - Accuracy: 0.9531
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0002
43
+ - train_batch_size: 16
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
+ - lr_scheduler_type: linear
48
+ - num_epochs: 4
49
+ - mixed_precision_training: Native AMP
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
54
+ |:-------------:|:------:|:----:|:---------------:|:--------:|
55
+ | 0.0677 | 1.5385 | 100 | 0.0555 | 0.9850 |
56
+ | 0.0148 | 3.0769 | 200 | 0.0133 | 1.0 |
57
+
58
+
59
+ ### Framework versions
60
+
61
+ - Transformers 4.41.2
62
+ - Pytorch 2.3.0+cu121
63
+ - Datasets 2.20.0
64
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "eval_accuracy": 0.953125,
4
+ "eval_loss": 0.19588422775268555,
5
+ "eval_runtime": 4.7533,
6
+ "eval_samples_per_second": 26.929,
7
+ "eval_steps_per_second": 3.366,
8
+ "total_flos": 3.205097416476426e+17,
9
+ "train_loss": 0.12496758367006595,
10
+ "train_runtime": 90.9233,
11
+ "train_samples_per_second": 45.489,
12
+ "train_steps_per_second": 2.86
13
+ }
config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "angular_leaf_spot",
13
+ "1": "bean_rust",
14
+ "2": "healthy"
15
+ },
16
+ "image_size": 224,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 3072,
19
+ "label2id": {
20
+ "angular_leaf_spot": "0",
21
+ "bean_rust": "1",
22
+ "healthy": "2"
23
+ },
24
+ "layer_norm_eps": 1e-12,
25
+ "model_type": "vit",
26
+ "num_attention_heads": 12,
27
+ "num_channels": 3,
28
+ "num_hidden_layers": 12,
29
+ "patch_size": 16,
30
+ "problem_type": "single_label_classification",
31
+ "qkv_bias": true,
32
+ "torch_dtype": "float32",
33
+ "transformers_version": "4.41.2"
34
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2ff39ae7faa909253e1fdd134bf7bba0ad18505b5d3df63c7f8454e07a1670fa
3
+ size 343227052
preprocessor_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_valid_processor_keys": [
3
+ "images",
4
+ "do_resize",
5
+ "size",
6
+ "resample",
7
+ "do_rescale",
8
+ "rescale_factor",
9
+ "do_normalize",
10
+ "image_mean",
11
+ "image_std",
12
+ "return_tensors",
13
+ "data_format",
14
+ "input_data_format"
15
+ ],
16
+ "do_normalize": true,
17
+ "do_rescale": true,
18
+ "do_resize": true,
19
+ "image_mean": [
20
+ 0.5,
21
+ 0.5,
22
+ 0.5
23
+ ],
24
+ "image_processor_type": "ViTFeatureExtractor",
25
+ "image_std": [
26
+ 0.5,
27
+ 0.5,
28
+ 0.5
29
+ ],
30
+ "resample": 2,
31
+ "rescale_factor": 0.00392156862745098,
32
+ "size": {
33
+ "height": 224,
34
+ "width": 224
35
+ }
36
+ }
runs/Jul08_06-44-18_030ec29a49e9/events.out.tfevents.1720422788.030ec29a49e9.843.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e5e137581c9e8854b1d856d92791ca8bbf5bbb2a39162b2e3a224ef3f2e5440
3
+ size 11323
runs/Jul08_06-44-18_030ec29a49e9/events.out.tfevents.1720423012.030ec29a49e9.843.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56769f806733cc48dfff88ead47cd8ef2bc059f3e0c02be868e83f2f6a94fa2f
3
+ size 411
test_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "eval_accuracy": 0.953125,
4
+ "eval_loss": 0.19588422775268555,
5
+ "eval_runtime": 4.7533,
6
+ "eval_samples_per_second": 26.929,
7
+ "eval_steps_per_second": 3.366
8
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "total_flos": 3.205097416476426e+17,
4
+ "train_loss": 0.12496758367006595,
5
+ "train_runtime": 90.9233,
6
+ "train_samples_per_second": 45.489,
7
+ "train_steps_per_second": 2.86
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,242 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.013332394883036613,
3
+ "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-200",
4
+ "epoch": 4.0,
5
+ "eval_steps": 100,
6
+ "global_step": 260,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.15384615384615385,
13
+ "grad_norm": 1.4326872825622559,
14
+ "learning_rate": 0.00019230769230769233,
15
+ "loss": 0.7835,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.3076923076923077,
20
+ "grad_norm": 1.9304112195968628,
21
+ "learning_rate": 0.00018461538461538463,
22
+ "loss": 0.356,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.46153846153846156,
27
+ "grad_norm": 3.0517399311065674,
28
+ "learning_rate": 0.00017692307692307693,
29
+ "loss": 0.2107,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.6153846153846154,
34
+ "grad_norm": 3.5913784503936768,
35
+ "learning_rate": 0.00016923076923076923,
36
+ "loss": 0.2222,
37
+ "step": 40
38
+ },
39
+ {
40
+ "epoch": 0.7692307692307693,
41
+ "grad_norm": 4.897500038146973,
42
+ "learning_rate": 0.00016153846153846155,
43
+ "loss": 0.2676,
44
+ "step": 50
45
+ },
46
+ {
47
+ "epoch": 0.9230769230769231,
48
+ "grad_norm": 3.6991374492645264,
49
+ "learning_rate": 0.00015384615384615385,
50
+ "loss": 0.3677,
51
+ "step": 60
52
+ },
53
+ {
54
+ "epoch": 1.0769230769230769,
55
+ "grad_norm": 0.5888941287994385,
56
+ "learning_rate": 0.00014615384615384615,
57
+ "loss": 0.0946,
58
+ "step": 70
59
+ },
60
+ {
61
+ "epoch": 1.2307692307692308,
62
+ "grad_norm": 5.136747360229492,
63
+ "learning_rate": 0.00013846153846153847,
64
+ "loss": 0.1049,
65
+ "step": 80
66
+ },
67
+ {
68
+ "epoch": 1.3846153846153846,
69
+ "grad_norm": 4.133493900299072,
70
+ "learning_rate": 0.00013076923076923077,
71
+ "loss": 0.0366,
72
+ "step": 90
73
+ },
74
+ {
75
+ "epoch": 1.5384615384615383,
76
+ "grad_norm": 5.138324737548828,
77
+ "learning_rate": 0.0001230769230769231,
78
+ "loss": 0.0677,
79
+ "step": 100
80
+ },
81
+ {
82
+ "epoch": 1.5384615384615383,
83
+ "eval_accuracy": 0.9849624060150376,
84
+ "eval_loss": 0.05553809180855751,
85
+ "eval_runtime": 1.2866,
86
+ "eval_samples_per_second": 103.375,
87
+ "eval_steps_per_second": 13.213,
88
+ "step": 100
89
+ },
90
+ {
91
+ "epoch": 1.6923076923076923,
92
+ "grad_norm": 0.1761542111635208,
93
+ "learning_rate": 0.00011538461538461538,
94
+ "loss": 0.1063,
95
+ "step": 110
96
+ },
97
+ {
98
+ "epoch": 1.8461538461538463,
99
+ "grad_norm": 3.5044777393341064,
100
+ "learning_rate": 0.0001076923076923077,
101
+ "loss": 0.1599,
102
+ "step": 120
103
+ },
104
+ {
105
+ "epoch": 2.0,
106
+ "grad_norm": 3.8694207668304443,
107
+ "learning_rate": 0.0001,
108
+ "loss": 0.1102,
109
+ "step": 130
110
+ },
111
+ {
112
+ "epoch": 2.1538461538461537,
113
+ "grad_norm": 0.2651481032371521,
114
+ "learning_rate": 9.230769230769232e-05,
115
+ "loss": 0.0379,
116
+ "step": 140
117
+ },
118
+ {
119
+ "epoch": 2.3076923076923075,
120
+ "grad_norm": 6.802685260772705,
121
+ "learning_rate": 8.461538461538461e-05,
122
+ "loss": 0.0465,
123
+ "step": 150
124
+ },
125
+ {
126
+ "epoch": 2.4615384615384617,
127
+ "grad_norm": 0.1000690683722496,
128
+ "learning_rate": 7.692307692307693e-05,
129
+ "loss": 0.0511,
130
+ "step": 160
131
+ },
132
+ {
133
+ "epoch": 2.6153846153846154,
134
+ "grad_norm": 1.7406809329986572,
135
+ "learning_rate": 6.923076923076924e-05,
136
+ "loss": 0.0206,
137
+ "step": 170
138
+ },
139
+ {
140
+ "epoch": 2.769230769230769,
141
+ "grad_norm": 0.2656266689300537,
142
+ "learning_rate": 6.153846153846155e-05,
143
+ "loss": 0.0656,
144
+ "step": 180
145
+ },
146
+ {
147
+ "epoch": 2.9230769230769234,
148
+ "grad_norm": 0.1580563485622406,
149
+ "learning_rate": 5.384615384615385e-05,
150
+ "loss": 0.0238,
151
+ "step": 190
152
+ },
153
+ {
154
+ "epoch": 3.076923076923077,
155
+ "grad_norm": 0.06784767657518387,
156
+ "learning_rate": 4.615384615384616e-05,
157
+ "loss": 0.0148,
158
+ "step": 200
159
+ },
160
+ {
161
+ "epoch": 3.076923076923077,
162
+ "eval_accuracy": 1.0,
163
+ "eval_loss": 0.013332394883036613,
164
+ "eval_runtime": 1.2533,
165
+ "eval_samples_per_second": 106.118,
166
+ "eval_steps_per_second": 13.564,
167
+ "step": 200
168
+ },
169
+ {
170
+ "epoch": 3.230769230769231,
171
+ "grad_norm": 0.059637218713760376,
172
+ "learning_rate": 3.846153846153846e-05,
173
+ "loss": 0.039,
174
+ "step": 210
175
+ },
176
+ {
177
+ "epoch": 3.3846153846153846,
178
+ "grad_norm": 0.0609792061150074,
179
+ "learning_rate": 3.0769230769230774e-05,
180
+ "loss": 0.012,
181
+ "step": 220
182
+ },
183
+ {
184
+ "epoch": 3.5384615384615383,
185
+ "grad_norm": 0.06473277509212494,
186
+ "learning_rate": 2.307692307692308e-05,
187
+ "loss": 0.0117,
188
+ "step": 230
189
+ },
190
+ {
191
+ "epoch": 3.6923076923076925,
192
+ "grad_norm": 0.06173160299658775,
193
+ "learning_rate": 1.5384615384615387e-05,
194
+ "loss": 0.0112,
195
+ "step": 240
196
+ },
197
+ {
198
+ "epoch": 3.8461538461538463,
199
+ "grad_norm": 0.05901215225458145,
200
+ "learning_rate": 7.692307692307694e-06,
201
+ "loss": 0.0109,
202
+ "step": 250
203
+ },
204
+ {
205
+ "epoch": 4.0,
206
+ "grad_norm": 0.05926782637834549,
207
+ "learning_rate": 0.0,
208
+ "loss": 0.0163,
209
+ "step": 260
210
+ },
211
+ {
212
+ "epoch": 4.0,
213
+ "step": 260,
214
+ "total_flos": 3.205097416476426e+17,
215
+ "train_loss": 0.12496758367006595,
216
+ "train_runtime": 90.9233,
217
+ "train_samples_per_second": 45.489,
218
+ "train_steps_per_second": 2.86
219
+ }
220
+ ],
221
+ "logging_steps": 10,
222
+ "max_steps": 260,
223
+ "num_input_tokens_seen": 0,
224
+ "num_train_epochs": 4,
225
+ "save_steps": 100,
226
+ "stateful_callbacks": {
227
+ "TrainerControl": {
228
+ "args": {
229
+ "should_epoch_stop": false,
230
+ "should_evaluate": false,
231
+ "should_log": false,
232
+ "should_save": true,
233
+ "should_training_stop": false
234
+ },
235
+ "attributes": {}
236
+ }
237
+ },
238
+ "total_flos": 3.205097416476426e+17,
239
+ "train_batch_size": 16,
240
+ "trial_name": null,
241
+ "trial_params": null
242
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:390874a77888dc21be66ec961698b59f9bf68ac399435f8c06b443c51c5e6cb1
3
+ size 5112