elvispresniy commited on
Commit
13e5233
1 Parent(s): 41e907d

End of training

Browse files
Files changed (5) hide show
  1. README.md +66 -0
  2. config.json +228 -0
  3. model.safetensors +3 -0
  4. preprocessor_config.json +22 -0
  5. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: google/vit-base-patch16-224-in21k
4
+ tags:
5
+ - generated_from_trainer
6
+ metrics:
7
+ - accuracy
8
+ model-index:
9
+ - name: vit-food101
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
+ # vit-food101
17
+
18
+ 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.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 0.4925
21
+ - Accuracy: 0.899
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0002
41
+ - train_batch_size: 64
42
+ - eval_batch_size: 8
43
+ - seed: 42
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: linear
46
+ - num_epochs: 4
47
+ - mixed_precision_training: Native AMP
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
52
+ |:-------------:|:------:|:----:|:---------------:|:--------:|
53
+ | 2.682 | 0.6369 | 100 | 2.5501 | 0.802 |
54
+ | 1.312 | 1.2739 | 200 | 1.3870 | 0.855 |
55
+ | 0.7605 | 1.9108 | 300 | 0.9167 | 0.862 |
56
+ | 0.3844 | 2.5478 | 400 | 0.6248 | 0.88 |
57
+ | 0.1957 | 3.1847 | 500 | 0.5220 | 0.896 |
58
+ | 0.1756 | 3.8217 | 600 | 0.4925 | 0.899 |
59
+
60
+
61
+ ### Framework versions
62
+
63
+ - Transformers 4.42.4
64
+ - Pytorch 2.3.1+cu121
65
+ - Datasets 2.20.0
66
+ - Tokenizers 0.19.1
config.json ADDED
@@ -0,0 +1,228 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "apple",
13
+ "1": "aquarium_fish",
14
+ "10": "bowl",
15
+ "11": "boy",
16
+ "12": "bridge",
17
+ "13": "bus",
18
+ "14": "butterfly",
19
+ "15": "camel",
20
+ "16": "can",
21
+ "17": "castle",
22
+ "18": "caterpillar",
23
+ "19": "cattle",
24
+ "2": "baby",
25
+ "20": "chair",
26
+ "21": "chimpanzee",
27
+ "22": "clock",
28
+ "23": "cloud",
29
+ "24": "cockroach",
30
+ "25": "couch",
31
+ "26": "cra",
32
+ "27": "crocodile",
33
+ "28": "cup",
34
+ "29": "dinosaur",
35
+ "3": "bear",
36
+ "30": "dolphin",
37
+ "31": "elephant",
38
+ "32": "flatfish",
39
+ "33": "forest",
40
+ "34": "fox",
41
+ "35": "girl",
42
+ "36": "hamster",
43
+ "37": "house",
44
+ "38": "kangaroo",
45
+ "39": "keyboard",
46
+ "4": "beaver",
47
+ "40": "lamp",
48
+ "41": "lawn_mower",
49
+ "42": "leopard",
50
+ "43": "lion",
51
+ "44": "lizard",
52
+ "45": "lobster",
53
+ "46": "man",
54
+ "47": "maple_tree",
55
+ "48": "motorcycle",
56
+ "49": "mountain",
57
+ "5": "bed",
58
+ "50": "mouse",
59
+ "51": "mushroom",
60
+ "52": "oak_tree",
61
+ "53": "orange",
62
+ "54": "orchid",
63
+ "55": "otter",
64
+ "56": "palm_tree",
65
+ "57": "pear",
66
+ "58": "pickup_truck",
67
+ "59": "pine_tree",
68
+ "6": "bee",
69
+ "60": "plain",
70
+ "61": "plate",
71
+ "62": "poppy",
72
+ "63": "porcupine",
73
+ "64": "possum",
74
+ "65": "rabbit",
75
+ "66": "raccoon",
76
+ "67": "ray",
77
+ "68": "road",
78
+ "69": "rocket",
79
+ "7": "beetle",
80
+ "70": "rose",
81
+ "71": "sea",
82
+ "72": "seal",
83
+ "73": "shark",
84
+ "74": "shrew",
85
+ "75": "skunk",
86
+ "76": "skyscraper",
87
+ "77": "snail",
88
+ "78": "snake",
89
+ "79": "spider",
90
+ "8": "bicycle",
91
+ "80": "squirrel",
92
+ "81": "streetcar",
93
+ "82": "sunflower",
94
+ "83": "sweet_pepper",
95
+ "84": "table",
96
+ "85": "tank",
97
+ "86": "telephone",
98
+ "87": "television",
99
+ "88": "tiger",
100
+ "89": "tractor",
101
+ "9": "bottle",
102
+ "90": "train",
103
+ "91": "trout",
104
+ "92": "tulip",
105
+ "93": "turtle",
106
+ "94": "wardrobe",
107
+ "95": "whale",
108
+ "96": "willow_tree",
109
+ "97": "wolf",
110
+ "98": "woman",
111
+ "99": "worm"
112
+ },
113
+ "image_size": 224,
114
+ "initializer_range": 0.02,
115
+ "intermediate_size": 3072,
116
+ "label2id": {
117
+ "apple": "0",
118
+ "aquarium_fish": "1",
119
+ "baby": "2",
120
+ "bear": "3",
121
+ "beaver": "4",
122
+ "bed": "5",
123
+ "bee": "6",
124
+ "beetle": "7",
125
+ "bicycle": "8",
126
+ "bottle": "9",
127
+ "bowl": "10",
128
+ "boy": "11",
129
+ "bridge": "12",
130
+ "bus": "13",
131
+ "butterfly": "14",
132
+ "camel": "15",
133
+ "can": "16",
134
+ "castle": "17",
135
+ "caterpillar": "18",
136
+ "cattle": "19",
137
+ "chair": "20",
138
+ "chimpanzee": "21",
139
+ "clock": "22",
140
+ "cloud": "23",
141
+ "cockroach": "24",
142
+ "couch": "25",
143
+ "cra": "26",
144
+ "crocodile": "27",
145
+ "cup": "28",
146
+ "dinosaur": "29",
147
+ "dolphin": "30",
148
+ "elephant": "31",
149
+ "flatfish": "32",
150
+ "forest": "33",
151
+ "fox": "34",
152
+ "girl": "35",
153
+ "hamster": "36",
154
+ "house": "37",
155
+ "kangaroo": "38",
156
+ "keyboard": "39",
157
+ "lamp": "40",
158
+ "lawn_mower": "41",
159
+ "leopard": "42",
160
+ "lion": "43",
161
+ "lizard": "44",
162
+ "lobster": "45",
163
+ "man": "46",
164
+ "maple_tree": "47",
165
+ "motorcycle": "48",
166
+ "mountain": "49",
167
+ "mouse": "50",
168
+ "mushroom": "51",
169
+ "oak_tree": "52",
170
+ "orange": "53",
171
+ "orchid": "54",
172
+ "otter": "55",
173
+ "palm_tree": "56",
174
+ "pear": "57",
175
+ "pickup_truck": "58",
176
+ "pine_tree": "59",
177
+ "plain": "60",
178
+ "plate": "61",
179
+ "poppy": "62",
180
+ "porcupine": "63",
181
+ "possum": "64",
182
+ "rabbit": "65",
183
+ "raccoon": "66",
184
+ "ray": "67",
185
+ "road": "68",
186
+ "rocket": "69",
187
+ "rose": "70",
188
+ "sea": "71",
189
+ "seal": "72",
190
+ "shark": "73",
191
+ "shrew": "74",
192
+ "skunk": "75",
193
+ "skyscraper": "76",
194
+ "snail": "77",
195
+ "snake": "78",
196
+ "spider": "79",
197
+ "squirrel": "80",
198
+ "streetcar": "81",
199
+ "sunflower": "82",
200
+ "sweet_pepper": "83",
201
+ "table": "84",
202
+ "tank": "85",
203
+ "telephone": "86",
204
+ "television": "87",
205
+ "tiger": "88",
206
+ "tractor": "89",
207
+ "train": "90",
208
+ "trout": "91",
209
+ "tulip": "92",
210
+ "turtle": "93",
211
+ "wardrobe": "94",
212
+ "whale": "95",
213
+ "willow_tree": "96",
214
+ "wolf": "97",
215
+ "woman": "98",
216
+ "worm": "99"
217
+ },
218
+ "layer_norm_eps": 1e-12,
219
+ "model_type": "vit",
220
+ "num_attention_heads": 12,
221
+ "num_channels": 3,
222
+ "num_hidden_layers": 12,
223
+ "patch_size": 16,
224
+ "problem_type": "single_label_classification",
225
+ "qkv_bias": true,
226
+ "torch_dtype": "float32",
227
+ "transformers_version": "4.42.4"
228
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f8247fc1bfd4993c348f650827718bcec383346671b151bfdd0b7dcbfdd53428
3
+ size 343525432
preprocessor_config.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": true,
3
+ "do_rescale": true,
4
+ "do_resize": true,
5
+ "image_mean": [
6
+ 0.5,
7
+ 0.5,
8
+ 0.5
9
+ ],
10
+ "image_processor_type": "ViTImageProcessor",
11
+ "image_std": [
12
+ 0.5,
13
+ 0.5,
14
+ 0.5
15
+ ],
16
+ "resample": 2,
17
+ "rescale_factor": 0.00392156862745098,
18
+ "size": {
19
+ "height": 224,
20
+ "width": 224
21
+ }
22
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9983f71b713b6c70d45d748db25b269da87a2fa566e1b9dc904ab42bd97d2a9
3
+ size 5112