dima806 commited on
Commit
4d1936a
1 Parent(s): 49a80ef

Upload folder using huggingface_hub

Browse files
checkpoint-4800/config.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "dima806/facial_age_image_detection",
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": "01",
13
+ "1": "02",
14
+ "2": "03",
15
+ "3": "04",
16
+ "4": "05",
17
+ "5": "06-07",
18
+ "6": "08-09",
19
+ "7": "10-12",
20
+ "8": "13-15",
21
+ "9": "16-20",
22
+ "10": "21-25",
23
+ "11": "26-30",
24
+ "12": "31-35",
25
+ "13": "36-40",
26
+ "14": "41-45",
27
+ "15": "46-50",
28
+ "16": "51-55",
29
+ "17": "56-60",
30
+ "18": "61-65",
31
+ "19": "66-70",
32
+ "20": "71-80",
33
+ "21": "81-90",
34
+ "22": "90+"
35
+ },
36
+ "image_size": 224,
37
+ "initializer_range": 0.02,
38
+ "intermediate_size": 3072,
39
+ "label2id": {
40
+ "01": 0,
41
+ "02": 1,
42
+ "03": 2,
43
+ "04": 3,
44
+ "05": 4,
45
+ "06-07": 5,
46
+ "08-09": 6,
47
+ "10-12": 7,
48
+ "13-15": 8,
49
+ "16-20": 9,
50
+ "21-25": 10,
51
+ "26-30": 11,
52
+ "31-35": 12,
53
+ "36-40": 13,
54
+ "41-45": 14,
55
+ "46-50": 15,
56
+ "51-55": 16,
57
+ "56-60": 17,
58
+ "61-65": 18,
59
+ "66-70": 19,
60
+ "71-80": 20,
61
+ "81-90": 21,
62
+ "90+": 22
63
+ },
64
+ "layer_norm_eps": 1e-12,
65
+ "model_type": "vit",
66
+ "num_attention_heads": 12,
67
+ "num_channels": 3,
68
+ "num_hidden_layers": 12,
69
+ "patch_size": 16,
70
+ "problem_type": "single_label_classification",
71
+ "qkv_bias": true,
72
+ "torch_dtype": "float32",
73
+ "transformers_version": "4.38.2"
74
+ }
checkpoint-4800/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ca355e4ab0436bd1f28b1391a4d956b8e70f091a4970fe45a99bd0c57eb8d48
3
+ size 343288572
checkpoint-4800/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:636f982860f05ed6489ed929f73e582710394915a4ae9db2dc4fbb4074adc5dd
3
+ size 686697605
checkpoint-4800/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
+ }
checkpoint-4800/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b63952a445a33fd5e643b4d8cf806228d069960e90d213d8424126151335d8f
3
+ size 14575
checkpoint-4800/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7966dfc291344112ca6820d5a4ac31b6e5fdb54e1e3983478f848ec06f323360
3
+ size 627
checkpoint-4800/trainer_state.json ADDED
@@ -0,0 +1,264 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.164973497390747,
3
+ "best_model_checkpoint": "facial_age_image_detection/checkpoint-4800",
4
+ "epoch": 20.0,
5
+ "eval_steps": 500,
6
+ "global_step": 4800,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "eval_accuracy": 0.6013097448929723,
14
+ "eval_loss": 1.182183861732483,
15
+ "eval_runtime": 106.5296,
16
+ "eval_samples_per_second": 96.039,
17
+ "eval_steps_per_second": 3.004,
18
+ "step": 240
19
+ },
20
+ {
21
+ "epoch": 2.0,
22
+ "eval_accuracy": 0.6016029713615483,
23
+ "eval_loss": 1.1812384128570557,
24
+ "eval_runtime": 103.6605,
25
+ "eval_samples_per_second": 98.697,
26
+ "eval_steps_per_second": 3.087,
27
+ "step": 480
28
+ },
29
+ {
30
+ "epoch": 2.08,
31
+ "grad_norm": 11.414525032043457,
32
+ "learning_rate": 4.526315789473684e-07,
33
+ "loss": 0.7517,
34
+ "step": 500
35
+ },
36
+ {
37
+ "epoch": 3.0,
38
+ "eval_accuracy": 0.5973023164891017,
39
+ "eval_loss": 1.1846318244934082,
40
+ "eval_runtime": 104.3279,
41
+ "eval_samples_per_second": 98.066,
42
+ "eval_steps_per_second": 3.067,
43
+ "step": 720
44
+ },
45
+ {
46
+ "epoch": 4.0,
47
+ "eval_accuracy": 0.6031668458606196,
48
+ "eval_loss": 1.178046464920044,
49
+ "eval_runtime": 105.5564,
50
+ "eval_samples_per_second": 96.924,
51
+ "eval_steps_per_second": 3.032,
52
+ "step": 960
53
+ },
54
+ {
55
+ "epoch": 4.17,
56
+ "grad_norm": 10.3371000289917,
57
+ "learning_rate": 4e-07,
58
+ "loss": 0.7404,
59
+ "step": 1000
60
+ },
61
+ {
62
+ "epoch": 5.0,
63
+ "eval_accuracy": 0.6057081419216108,
64
+ "eval_loss": 1.1763097047805786,
65
+ "eval_runtime": 104.8017,
66
+ "eval_samples_per_second": 97.622,
67
+ "eval_steps_per_second": 3.053,
68
+ "step": 1200
69
+ },
70
+ {
71
+ "epoch": 6.0,
72
+ "eval_accuracy": 0.6061968527025706,
73
+ "eval_loss": 1.173648715019226,
74
+ "eval_runtime": 104.8949,
75
+ "eval_samples_per_second": 97.536,
76
+ "eval_steps_per_second": 3.051,
77
+ "step": 1440
78
+ },
79
+ {
80
+ "epoch": 6.25,
81
+ "grad_norm": 11.105679512023926,
82
+ "learning_rate": 3.4736842105263157e-07,
83
+ "loss": 0.7343,
84
+ "step": 1500
85
+ },
86
+ {
87
+ "epoch": 7.0,
88
+ "eval_accuracy": 0.6017984556739322,
89
+ "eval_loss": 1.175057053565979,
90
+ "eval_runtime": 104.1925,
91
+ "eval_samples_per_second": 98.193,
92
+ "eval_steps_per_second": 3.071,
93
+ "step": 1680
94
+ },
95
+ {
96
+ "epoch": 8.0,
97
+ "eval_accuracy": 0.6026781350796598,
98
+ "eval_loss": 1.1735320091247559,
99
+ "eval_runtime": 103.0462,
100
+ "eval_samples_per_second": 99.286,
101
+ "eval_steps_per_second": 3.105,
102
+ "step": 1920
103
+ },
104
+ {
105
+ "epoch": 8.33,
106
+ "grad_norm": 15.928546905517578,
107
+ "learning_rate": 2.947368421052631e-07,
108
+ "loss": 0.7275,
109
+ "step": 2000
110
+ },
111
+ {
112
+ "epoch": 9.0,
113
+ "eval_accuracy": 0.6044374938911152,
114
+ "eval_loss": 1.1707645654678345,
115
+ "eval_runtime": 103.3329,
116
+ "eval_samples_per_second": 99.01,
117
+ "eval_steps_per_second": 3.097,
118
+ "step": 2160
119
+ },
120
+ {
121
+ "epoch": 10.0,
122
+ "eval_accuracy": 0.6085426644511778,
123
+ "eval_loss": 1.169615387916565,
124
+ "eval_runtime": 104.1315,
125
+ "eval_samples_per_second": 98.251,
126
+ "eval_steps_per_second": 3.073,
127
+ "step": 2400
128
+ },
129
+ {
130
+ "epoch": 10.42,
131
+ "grad_norm": 15.35641098022461,
132
+ "learning_rate": 2.4210526315789473e-07,
133
+ "loss": 0.7201,
134
+ "step": 2500
135
+ },
136
+ {
137
+ "epoch": 11.0,
138
+ "eval_accuracy": 0.6064900791711465,
139
+ "eval_loss": 1.1700137853622437,
140
+ "eval_runtime": 106.578,
141
+ "eval_samples_per_second": 95.995,
142
+ "eval_steps_per_second": 3.002,
143
+ "step": 2640
144
+ },
145
+ {
146
+ "epoch": 12.0,
147
+ "eval_accuracy": 0.6087381487635617,
148
+ "eval_loss": 1.1673282384872437,
149
+ "eval_runtime": 102.521,
150
+ "eval_samples_per_second": 99.794,
151
+ "eval_steps_per_second": 3.121,
152
+ "step": 2880
153
+ },
154
+ {
155
+ "epoch": 12.5,
156
+ "grad_norm": 10.890913963317871,
157
+ "learning_rate": 1.8947368421052632e-07,
158
+ "loss": 0.7144,
159
+ "step": 3000
160
+ },
161
+ {
162
+ "epoch": 13.0,
163
+ "eval_accuracy": 0.6035578144853876,
164
+ "eval_loss": 1.1682850122451782,
165
+ "eval_runtime": 103.3961,
166
+ "eval_samples_per_second": 98.95,
167
+ "eval_steps_per_second": 3.095,
168
+ "step": 3120
169
+ },
170
+ {
171
+ "epoch": 14.0,
172
+ "eval_accuracy": 0.6070765321082983,
173
+ "eval_loss": 1.1674185991287231,
174
+ "eval_runtime": 103.2219,
175
+ "eval_samples_per_second": 99.117,
176
+ "eval_steps_per_second": 3.1,
177
+ "step": 3360
178
+ },
179
+ {
180
+ "epoch": 14.58,
181
+ "grad_norm": 11.558945655822754,
182
+ "learning_rate": 1.368421052631579e-07,
183
+ "loss": 0.7137,
184
+ "step": 3500
185
+ },
186
+ {
187
+ "epoch": 15.0,
188
+ "eval_accuracy": 0.6073697585768743,
189
+ "eval_loss": 1.1667180061340332,
190
+ "eval_runtime": 103.2403,
191
+ "eval_samples_per_second": 99.099,
192
+ "eval_steps_per_second": 3.1,
193
+ "step": 3600
194
+ },
195
+ {
196
+ "epoch": 16.0,
197
+ "eval_accuracy": 0.6068810477959143,
198
+ "eval_loss": 1.166371464729309,
199
+ "eval_runtime": 103.7779,
200
+ "eval_samples_per_second": 98.586,
201
+ "eval_steps_per_second": 3.084,
202
+ "step": 3840
203
+ },
204
+ {
205
+ "epoch": 16.67,
206
+ "grad_norm": 15.126260757446289,
207
+ "learning_rate": 8.421052631578947e-08,
208
+ "loss": 0.7093,
209
+ "step": 4000
210
+ },
211
+ {
212
+ "epoch": 17.0,
213
+ "eval_accuracy": 0.6044374938911152,
214
+ "eval_loss": 1.1674280166625977,
215
+ "eval_runtime": 102.8109,
216
+ "eval_samples_per_second": 99.513,
217
+ "eval_steps_per_second": 3.113,
218
+ "step": 4080
219
+ },
220
+ {
221
+ "epoch": 18.0,
222
+ "eval_accuracy": 0.6042420095787313,
223
+ "eval_loss": 1.1664873361587524,
224
+ "eval_runtime": 102.8553,
225
+ "eval_samples_per_second": 99.47,
226
+ "eval_steps_per_second": 3.111,
227
+ "step": 4320
228
+ },
229
+ {
230
+ "epoch": 18.75,
231
+ "grad_norm": 14.217340469360352,
232
+ "learning_rate": 3.157894736842105e-08,
233
+ "loss": 0.7035,
234
+ "step": 4500
235
+ },
236
+ {
237
+ "epoch": 19.0,
238
+ "eval_accuracy": 0.6062945948587626,
239
+ "eval_loss": 1.1650714874267578,
240
+ "eval_runtime": 103.4609,
241
+ "eval_samples_per_second": 98.888,
242
+ "eval_steps_per_second": 3.093,
243
+ "step": 4560
244
+ },
245
+ {
246
+ "epoch": 20.0,
247
+ "eval_accuracy": 0.6068810477959143,
248
+ "eval_loss": 1.164973497390747,
249
+ "eval_runtime": 102.8512,
250
+ "eval_samples_per_second": 99.474,
251
+ "eval_steps_per_second": 3.111,
252
+ "step": 4800
253
+ }
254
+ ],
255
+ "logging_steps": 500,
256
+ "max_steps": 4800,
257
+ "num_input_tokens_seen": 0,
258
+ "num_train_epochs": 20,
259
+ "save_steps": 500,
260
+ "total_flos": 2.3786767829022114e+19,
261
+ "train_batch_size": 64,
262
+ "trial_name": null,
263
+ "trial_params": null
264
+ }
checkpoint-4800/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1720d13f8074df3fb08ee60b875a9df5a8b4161f2f31054e04a3b9fdc006e59a
3
+ size 4411
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0cecd28b4d92743518c8f4fb11713818229a881dd452670d73ff23b062afd1f7
3
  size 343288572
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ca355e4ab0436bd1f28b1391a4d956b8e70f091a4970fe45a99bd0c57eb8d48
3
  size 343288572