refactor: Organize model artifacts into 'models' directory
Browse files- vit-deepfake-model/checkpoint-1400/optimizer.pt → models/onnx/verichain-model.onnx +2 -2
- {vit-deepfake-model/checkpoint-1400 → models/vit-deepfake-model}/config.json +0 -0
- {vit-deepfake-model/checkpoint-1400 → models/vit-deepfake-model}/model.safetensors +1 -1
- {vit-deepfake-model/checkpoint-1400 → models/vit-deepfake-model}/preprocessor_config.json +0 -0
- {vit-deepfake-model/checkpoint-1400 → models/vit-deepfake-model}/training_args.bin +1 -1
- vit-deepfake-model/checkpoint-1400/rng_state.pth +0 -3
- vit-deepfake-model/checkpoint-1400/scheduler.pt +0 -3
- vit-deepfake-model/checkpoint-1400/trainer_state.json +0 -356
- vit-deepfake-model/checkpoint-1500/config.json +0 -35
- vit-deepfake-model/checkpoint-1500/model.safetensors +0 -3
- vit-deepfake-model/checkpoint-1500/optimizer.pt +0 -3
- vit-deepfake-model/checkpoint-1500/preprocessor_config.json +0 -23
- vit-deepfake-model/checkpoint-1500/rng_state.pth +0 -3
- vit-deepfake-model/checkpoint-1500/scheduler.pt +0 -3
- vit-deepfake-model/checkpoint-1500/trainer_state.json +0 -379
- vit-deepfake-model/checkpoint-1500/training_args.bin +0 -3
- vit-deepfake-model/config.json +0 -35
- vit-deepfake-model/confusion_matrix.png +0 -0
- vit-deepfake-model/model.safetensors +0 -3
- vit-deepfake-model/preprocessor_config.json +0 -23
- vit-deepfake-model/training_args.bin +0 -3
vit-deepfake-model/checkpoint-1400/optimizer.pt → models/onnx/verichain-model.onnx
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5030412cb31b0e2d7638635a95b712e2d1758041dc7a93cb4174975cf6826979
|
| 3 |
+
size 343480839
|
{vit-deepfake-model/checkpoint-1400 → models/vit-deepfake-model}/config.json
RENAMED
|
File without changes
|
{vit-deepfake-model/checkpoint-1400 → models/vit-deepfake-model}/model.safetensors
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343227052
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ac83bb37fd6c480aed3ef08619da92f1db169a249d1c4ea3cb9bb51fc8d2700
|
| 3 |
size 343227052
|
{vit-deepfake-model/checkpoint-1400 → models/vit-deepfake-model}/preprocessor_config.json
RENAMED
|
File without changes
|
{vit-deepfake-model/checkpoint-1400 → models/vit-deepfake-model}/training_args.bin
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5304
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b3bc803f87de7508b215673e673887458b1725038141333e73bfa4366fefba1c
|
| 3 |
size 5304
|
vit-deepfake-model/checkpoint-1400/rng_state.pth
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:78ab10d5c72c33a4113dd22d7cf9284b386d25e540228c5336f7eabfb16f3120
|
| 3 |
-
size 14244
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1400/scheduler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:ebaac472f8b1e446d959bf24a61a37b4b2399abf1223de42df5354dda678cfbc
|
| 3 |
-
size 1064
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1400/trainer_state.json
DELETED
|
@@ -1,356 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": 1300,
|
| 3 |
-
"best_metric": 0.019463825970888138,
|
| 4 |
-
"best_model_checkpoint": "./model/vit-deepfake-model/checkpoint-1200",
|
| 5 |
-
"epoch": 2.8,
|
| 6 |
-
"eval_steps": 100,
|
| 7 |
-
"global_step": 1400,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 0.1,
|
| 14 |
-
"grad_norm": 1.4394938945770264,
|
| 15 |
-
"learning_rate": 1.934666666666667e-05,
|
| 16 |
-
"loss": 0.8097,
|
| 17 |
-
"step": 50
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.2,
|
| 21 |
-
"grad_norm": 0.7798331379890442,
|
| 22 |
-
"learning_rate": 1.8680000000000004e-05,
|
| 23 |
-
"loss": 0.2637,
|
| 24 |
-
"step": 100
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.2,
|
| 28 |
-
"eval_accuracy": 0.991,
|
| 29 |
-
"eval_loss": 0.15913085639476776,
|
| 30 |
-
"eval_runtime": 42.0723,
|
| 31 |
-
"eval_samples_per_second": 47.537,
|
| 32 |
-
"eval_steps_per_second": 5.942,
|
| 33 |
-
"step": 100
|
| 34 |
-
},
|
| 35 |
-
{
|
| 36 |
-
"epoch": 0.3,
|
| 37 |
-
"grad_norm": 0.3808038532733917,
|
| 38 |
-
"learning_rate": 1.8013333333333333e-05,
|
| 39 |
-
"loss": 0.112,
|
| 40 |
-
"step": 150
|
| 41 |
-
},
|
| 42 |
-
{
|
| 43 |
-
"epoch": 0.4,
|
| 44 |
-
"grad_norm": 1.0245436429977417,
|
| 45 |
-
"learning_rate": 1.7346666666666668e-05,
|
| 46 |
-
"loss": 0.09,
|
| 47 |
-
"step": 200
|
| 48 |
-
},
|
| 49 |
-
{
|
| 50 |
-
"epoch": 0.4,
|
| 51 |
-
"eval_accuracy": 0.9895,
|
| 52 |
-
"eval_loss": 0.08769073337316513,
|
| 53 |
-
"eval_runtime": 43.4637,
|
| 54 |
-
"eval_samples_per_second": 46.015,
|
| 55 |
-
"eval_steps_per_second": 5.752,
|
| 56 |
-
"step": 200
|
| 57 |
-
},
|
| 58 |
-
{
|
| 59 |
-
"epoch": 0.5,
|
| 60 |
-
"grad_norm": 0.33664369583129883,
|
| 61 |
-
"learning_rate": 1.668e-05,
|
| 62 |
-
"loss": 0.0704,
|
| 63 |
-
"step": 250
|
| 64 |
-
},
|
| 65 |
-
{
|
| 66 |
-
"epoch": 0.6,
|
| 67 |
-
"grad_norm": 3.125309944152832,
|
| 68 |
-
"learning_rate": 1.6013333333333335e-05,
|
| 69 |
-
"loss": 0.0583,
|
| 70 |
-
"step": 300
|
| 71 |
-
},
|
| 72 |
-
{
|
| 73 |
-
"epoch": 0.6,
|
| 74 |
-
"eval_accuracy": 0.998,
|
| 75 |
-
"eval_loss": 0.04995392635464668,
|
| 76 |
-
"eval_runtime": 42.7949,
|
| 77 |
-
"eval_samples_per_second": 46.735,
|
| 78 |
-
"eval_steps_per_second": 5.842,
|
| 79 |
-
"step": 300
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.7,
|
| 83 |
-
"grad_norm": 0.19215664267539978,
|
| 84 |
-
"learning_rate": 1.5346666666666667e-05,
|
| 85 |
-
"loss": 0.0414,
|
| 86 |
-
"step": 350
|
| 87 |
-
},
|
| 88 |
-
{
|
| 89 |
-
"epoch": 0.8,
|
| 90 |
-
"grad_norm": 0.1444077342748642,
|
| 91 |
-
"learning_rate": 1.4680000000000002e-05,
|
| 92 |
-
"loss": 0.0454,
|
| 93 |
-
"step": 400
|
| 94 |
-
},
|
| 95 |
-
{
|
| 96 |
-
"epoch": 0.8,
|
| 97 |
-
"eval_accuracy": 0.998,
|
| 98 |
-
"eval_loss": 0.03876766189932823,
|
| 99 |
-
"eval_runtime": 42.439,
|
| 100 |
-
"eval_samples_per_second": 47.126,
|
| 101 |
-
"eval_steps_per_second": 5.891,
|
| 102 |
-
"step": 400
|
| 103 |
-
},
|
| 104 |
-
{
|
| 105 |
-
"epoch": 0.9,
|
| 106 |
-
"grad_norm": 0.13351820409297943,
|
| 107 |
-
"learning_rate": 1.4013333333333334e-05,
|
| 108 |
-
"loss": 0.0351,
|
| 109 |
-
"step": 450
|
| 110 |
-
},
|
| 111 |
-
{
|
| 112 |
-
"epoch": 1.0,
|
| 113 |
-
"grad_norm": 1.3633350133895874,
|
| 114 |
-
"learning_rate": 1.3346666666666667e-05,
|
| 115 |
-
"loss": 0.029,
|
| 116 |
-
"step": 500
|
| 117 |
-
},
|
| 118 |
-
{
|
| 119 |
-
"epoch": 1.0,
|
| 120 |
-
"eval_accuracy": 0.9985,
|
| 121 |
-
"eval_loss": 0.03332938253879547,
|
| 122 |
-
"eval_runtime": 42.2001,
|
| 123 |
-
"eval_samples_per_second": 47.393,
|
| 124 |
-
"eval_steps_per_second": 5.924,
|
| 125 |
-
"step": 500
|
| 126 |
-
},
|
| 127 |
-
{
|
| 128 |
-
"epoch": 1.1,
|
| 129 |
-
"grad_norm": 0.12323179095983505,
|
| 130 |
-
"learning_rate": 1.268e-05,
|
| 131 |
-
"loss": 0.0249,
|
| 132 |
-
"step": 550
|
| 133 |
-
},
|
| 134 |
-
{
|
| 135 |
-
"epoch": 1.2,
|
| 136 |
-
"grad_norm": 0.12278539687395096,
|
| 137 |
-
"learning_rate": 1.2013333333333334e-05,
|
| 138 |
-
"loss": 0.0301,
|
| 139 |
-
"step": 600
|
| 140 |
-
},
|
| 141 |
-
{
|
| 142 |
-
"epoch": 1.2,
|
| 143 |
-
"eval_accuracy": 0.9975,
|
| 144 |
-
"eval_loss": 0.031327877193689346,
|
| 145 |
-
"eval_runtime": 42.3539,
|
| 146 |
-
"eval_samples_per_second": 47.221,
|
| 147 |
-
"eval_steps_per_second": 5.903,
|
| 148 |
-
"step": 600
|
| 149 |
-
},
|
| 150 |
-
{
|
| 151 |
-
"epoch": 1.3,
|
| 152 |
-
"grad_norm": 0.12111877650022507,
|
| 153 |
-
"learning_rate": 1.1346666666666668e-05,
|
| 154 |
-
"loss": 0.0322,
|
| 155 |
-
"step": 650
|
| 156 |
-
},
|
| 157 |
-
{
|
| 158 |
-
"epoch": 1.4,
|
| 159 |
-
"grad_norm": 0.10371693968772888,
|
| 160 |
-
"learning_rate": 1.0680000000000001e-05,
|
| 161 |
-
"loss": 0.0243,
|
| 162 |
-
"step": 700
|
| 163 |
-
},
|
| 164 |
-
{
|
| 165 |
-
"epoch": 1.4,
|
| 166 |
-
"eval_accuracy": 0.997,
|
| 167 |
-
"eval_loss": 0.029479531571269035,
|
| 168 |
-
"eval_runtime": 42.2575,
|
| 169 |
-
"eval_samples_per_second": 47.329,
|
| 170 |
-
"eval_steps_per_second": 5.916,
|
| 171 |
-
"step": 700
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 1.5,
|
| 175 |
-
"grad_norm": 0.09562443196773529,
|
| 176 |
-
"learning_rate": 1.0013333333333335e-05,
|
| 177 |
-
"loss": 0.026,
|
| 178 |
-
"step": 750
|
| 179 |
-
},
|
| 180 |
-
{
|
| 181 |
-
"epoch": 1.6,
|
| 182 |
-
"grad_norm": 0.09563508629798889,
|
| 183 |
-
"learning_rate": 9.346666666666666e-06,
|
| 184 |
-
"loss": 0.0181,
|
| 185 |
-
"step": 800
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 1.6,
|
| 189 |
-
"eval_accuracy": 0.997,
|
| 190 |
-
"eval_loss": 0.027446668595075607,
|
| 191 |
-
"eval_runtime": 42.1658,
|
| 192 |
-
"eval_samples_per_second": 47.432,
|
| 193 |
-
"eval_steps_per_second": 5.929,
|
| 194 |
-
"step": 800
|
| 195 |
-
},
|
| 196 |
-
{
|
| 197 |
-
"epoch": 1.7,
|
| 198 |
-
"grad_norm": 0.08600907772779465,
|
| 199 |
-
"learning_rate": 8.68e-06,
|
| 200 |
-
"loss": 0.0252,
|
| 201 |
-
"step": 850
|
| 202 |
-
},
|
| 203 |
-
{
|
| 204 |
-
"epoch": 1.8,
|
| 205 |
-
"grad_norm": 0.10505673289299011,
|
| 206 |
-
"learning_rate": 8.013333333333333e-06,
|
| 207 |
-
"loss": 0.0167,
|
| 208 |
-
"step": 900
|
| 209 |
-
},
|
| 210 |
-
{
|
| 211 |
-
"epoch": 1.8,
|
| 212 |
-
"eval_accuracy": 0.997,
|
| 213 |
-
"eval_loss": 0.02535245567560196,
|
| 214 |
-
"eval_runtime": 42.2891,
|
| 215 |
-
"eval_samples_per_second": 47.294,
|
| 216 |
-
"eval_steps_per_second": 5.912,
|
| 217 |
-
"step": 900
|
| 218 |
-
},
|
| 219 |
-
{
|
| 220 |
-
"epoch": 1.9,
|
| 221 |
-
"grad_norm": 0.08181212097406387,
|
| 222 |
-
"learning_rate": 7.346666666666668e-06,
|
| 223 |
-
"loss": 0.016,
|
| 224 |
-
"step": 950
|
| 225 |
-
},
|
| 226 |
-
{
|
| 227 |
-
"epoch": 2.0,
|
| 228 |
-
"grad_norm": 0.07987715303897858,
|
| 229 |
-
"learning_rate": 6.680000000000001e-06,
|
| 230 |
-
"loss": 0.0159,
|
| 231 |
-
"step": 1000
|
| 232 |
-
},
|
| 233 |
-
{
|
| 234 |
-
"epoch": 2.0,
|
| 235 |
-
"eval_accuracy": 0.9955,
|
| 236 |
-
"eval_loss": 0.028334029018878937,
|
| 237 |
-
"eval_runtime": 42.9091,
|
| 238 |
-
"eval_samples_per_second": 46.61,
|
| 239 |
-
"eval_steps_per_second": 5.826,
|
| 240 |
-
"step": 1000
|
| 241 |
-
},
|
| 242 |
-
{
|
| 243 |
-
"epoch": 2.1,
|
| 244 |
-
"grad_norm": 17.688356399536133,
|
| 245 |
-
"learning_rate": 6.013333333333335e-06,
|
| 246 |
-
"loss": 0.0191,
|
| 247 |
-
"step": 1050
|
| 248 |
-
},
|
| 249 |
-
{
|
| 250 |
-
"epoch": 2.2,
|
| 251 |
-
"grad_norm": 0.07392571121454239,
|
| 252 |
-
"learning_rate": 5.346666666666667e-06,
|
| 253 |
-
"loss": 0.0154,
|
| 254 |
-
"step": 1100
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 2.2,
|
| 258 |
-
"eval_accuracy": 0.9985,
|
| 259 |
-
"eval_loss": 0.021651165559887886,
|
| 260 |
-
"eval_runtime": 42.7586,
|
| 261 |
-
"eval_samples_per_second": 46.774,
|
| 262 |
-
"eval_steps_per_second": 5.847,
|
| 263 |
-
"step": 1100
|
| 264 |
-
},
|
| 265 |
-
{
|
| 266 |
-
"epoch": 2.3,
|
| 267 |
-
"grad_norm": 0.09023883938789368,
|
| 268 |
-
"learning_rate": 4.680000000000001e-06,
|
| 269 |
-
"loss": 0.014,
|
| 270 |
-
"step": 1150
|
| 271 |
-
},
|
| 272 |
-
{
|
| 273 |
-
"epoch": 2.4,
|
| 274 |
-
"grad_norm": 0.075103759765625,
|
| 275 |
-
"learning_rate": 4.013333333333334e-06,
|
| 276 |
-
"loss": 0.0155,
|
| 277 |
-
"step": 1200
|
| 278 |
-
},
|
| 279 |
-
{
|
| 280 |
-
"epoch": 2.4,
|
| 281 |
-
"eval_accuracy": 0.998,
|
| 282 |
-
"eval_loss": 0.021380621939897537,
|
| 283 |
-
"eval_runtime": 42.7122,
|
| 284 |
-
"eval_samples_per_second": 46.825,
|
| 285 |
-
"eval_steps_per_second": 5.853,
|
| 286 |
-
"step": 1200
|
| 287 |
-
},
|
| 288 |
-
{
|
| 289 |
-
"epoch": 2.5,
|
| 290 |
-
"grad_norm": 0.07060286402702332,
|
| 291 |
-
"learning_rate": 3.346666666666667e-06,
|
| 292 |
-
"loss": 0.0135,
|
| 293 |
-
"step": 1250
|
| 294 |
-
},
|
| 295 |
-
{
|
| 296 |
-
"epoch": 2.6,
|
| 297 |
-
"grad_norm": 0.06728706508874893,
|
| 298 |
-
"learning_rate": 2.68e-06,
|
| 299 |
-
"loss": 0.0133,
|
| 300 |
-
"step": 1300
|
| 301 |
-
},
|
| 302 |
-
{
|
| 303 |
-
"epoch": 2.6,
|
| 304 |
-
"eval_accuracy": 0.9985,
|
| 305 |
-
"eval_loss": 0.019463825970888138,
|
| 306 |
-
"eval_runtime": 42.4277,
|
| 307 |
-
"eval_samples_per_second": 47.139,
|
| 308 |
-
"eval_steps_per_second": 5.892,
|
| 309 |
-
"step": 1300
|
| 310 |
-
},
|
| 311 |
-
{
|
| 312 |
-
"epoch": 2.7,
|
| 313 |
-
"grad_norm": 0.07170647382736206,
|
| 314 |
-
"learning_rate": 2.0133333333333337e-06,
|
| 315 |
-
"loss": 0.0139,
|
| 316 |
-
"step": 1350
|
| 317 |
-
},
|
| 318 |
-
{
|
| 319 |
-
"epoch": 2.8,
|
| 320 |
-
"grad_norm": 0.07239634543657303,
|
| 321 |
-
"learning_rate": 1.3466666666666668e-06,
|
| 322 |
-
"loss": 0.0129,
|
| 323 |
-
"step": 1400
|
| 324 |
-
},
|
| 325 |
-
{
|
| 326 |
-
"epoch": 2.8,
|
| 327 |
-
"eval_accuracy": 0.9985,
|
| 328 |
-
"eval_loss": 0.019522910937666893,
|
| 329 |
-
"eval_runtime": 42.425,
|
| 330 |
-
"eval_samples_per_second": 47.142,
|
| 331 |
-
"eval_steps_per_second": 5.893,
|
| 332 |
-
"step": 1400
|
| 333 |
-
}
|
| 334 |
-
],
|
| 335 |
-
"logging_steps": 50,
|
| 336 |
-
"max_steps": 1500,
|
| 337 |
-
"num_input_tokens_seen": 0,
|
| 338 |
-
"num_train_epochs": 3,
|
| 339 |
-
"save_steps": 200,
|
| 340 |
-
"stateful_callbacks": {
|
| 341 |
-
"TrainerControl": {
|
| 342 |
-
"args": {
|
| 343 |
-
"should_epoch_stop": false,
|
| 344 |
-
"should_evaluate": false,
|
| 345 |
-
"should_log": false,
|
| 346 |
-
"should_save": true,
|
| 347 |
-
"should_training_stop": false
|
| 348 |
-
},
|
| 349 |
-
"attributes": {}
|
| 350 |
-
}
|
| 351 |
-
},
|
| 352 |
-
"total_flos": 1.73568113960926e+18,
|
| 353 |
-
"train_batch_size": 16,
|
| 354 |
-
"trial_name": null,
|
| 355 |
-
"trial_params": null
|
| 356 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1500/config.json
DELETED
|
@@ -1,35 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"ViTForImageClassification"
|
| 4 |
-
],
|
| 5 |
-
"attention_probs_dropout_prob": 0.0,
|
| 6 |
-
"encoder_stride": 16,
|
| 7 |
-
"hidden_act": "gelu",
|
| 8 |
-
"hidden_dropout_prob": 0.0,
|
| 9 |
-
"hidden_size": 768,
|
| 10 |
-
"id2label": {
|
| 11 |
-
"0": "AI Generated",
|
| 12 |
-
"1": "Deepfake",
|
| 13 |
-
"2": "Real"
|
| 14 |
-
},
|
| 15 |
-
"image_size": 224,
|
| 16 |
-
"initializer_range": 0.02,
|
| 17 |
-
"intermediate_size": 3072,
|
| 18 |
-
"label2id": {
|
| 19 |
-
"AI Generated": 0,
|
| 20 |
-
"Deepfake": 1,
|
| 21 |
-
"Real": 2
|
| 22 |
-
},
|
| 23 |
-
"layer_norm_eps": 1e-12,
|
| 24 |
-
"model_type": "vit",
|
| 25 |
-
"num_attention_heads": 12,
|
| 26 |
-
"num_channels": 3,
|
| 27 |
-
"num_hidden_layers": 12,
|
| 28 |
-
"patch_size": 16,
|
| 29 |
-
"pooler_act": "tanh",
|
| 30 |
-
"pooler_output_size": 768,
|
| 31 |
-
"problem_type": "single_label_classification",
|
| 32 |
-
"qkv_bias": true,
|
| 33 |
-
"torch_dtype": "float32",
|
| 34 |
-
"transformers_version": "4.53.1"
|
| 35 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1500/model.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:eee4f6e7a8f62cfa2027345fbe75408298bc7a2f9db2dca567b018e7f0434787
|
| 3 |
-
size 343227052
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1500/optimizer.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:8d809e55d4a7da39b568455aa327e15d0b8d13da0aad4421fe377f59eb683253
|
| 3 |
-
size 686575034
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1500/preprocessor_config.json
DELETED
|
@@ -1,23 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"do_convert_rgb": null,
|
| 3 |
-
"do_normalize": true,
|
| 4 |
-
"do_rescale": true,
|
| 5 |
-
"do_resize": true,
|
| 6 |
-
"image_mean": [
|
| 7 |
-
0.5,
|
| 8 |
-
0.5,
|
| 9 |
-
0.5
|
| 10 |
-
],
|
| 11 |
-
"image_processor_type": "ViTImageProcessor",
|
| 12 |
-
"image_std": [
|
| 13 |
-
0.5,
|
| 14 |
-
0.5,
|
| 15 |
-
0.5
|
| 16 |
-
],
|
| 17 |
-
"resample": 2,
|
| 18 |
-
"rescale_factor": 0.00392156862745098,
|
| 19 |
-
"size": {
|
| 20 |
-
"height": 224,
|
| 21 |
-
"width": 224
|
| 22 |
-
}
|
| 23 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1500/rng_state.pth
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:783b6d49e27cd20232888a7613e8b332146f05c8aba8a0a87315f7c748889c20
|
| 3 |
-
size 14244
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1500/scheduler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:aeef384dca819c9b44bc575058a43da9e90e28ca7bdc6079aaf024944abda117
|
| 3 |
-
size 1064
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1500/trainer_state.json
DELETED
|
@@ -1,379 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": 1500,
|
| 3 |
-
"best_metric": 0.0190621018409729,
|
| 4 |
-
"best_model_checkpoint": "./model/vit-deepfake-model/checkpoint-1500",
|
| 5 |
-
"epoch": 3.0,
|
| 6 |
-
"eval_steps": 100,
|
| 7 |
-
"global_step": 1500,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 0.1,
|
| 14 |
-
"grad_norm": 1.4394938945770264,
|
| 15 |
-
"learning_rate": 1.934666666666667e-05,
|
| 16 |
-
"loss": 0.8097,
|
| 17 |
-
"step": 50
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.2,
|
| 21 |
-
"grad_norm": 0.7798331379890442,
|
| 22 |
-
"learning_rate": 1.8680000000000004e-05,
|
| 23 |
-
"loss": 0.2637,
|
| 24 |
-
"step": 100
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.2,
|
| 28 |
-
"eval_accuracy": 0.991,
|
| 29 |
-
"eval_loss": 0.15913085639476776,
|
| 30 |
-
"eval_runtime": 42.0723,
|
| 31 |
-
"eval_samples_per_second": 47.537,
|
| 32 |
-
"eval_steps_per_second": 5.942,
|
| 33 |
-
"step": 100
|
| 34 |
-
},
|
| 35 |
-
{
|
| 36 |
-
"epoch": 0.3,
|
| 37 |
-
"grad_norm": 0.3808038532733917,
|
| 38 |
-
"learning_rate": 1.8013333333333333e-05,
|
| 39 |
-
"loss": 0.112,
|
| 40 |
-
"step": 150
|
| 41 |
-
},
|
| 42 |
-
{
|
| 43 |
-
"epoch": 0.4,
|
| 44 |
-
"grad_norm": 1.0245436429977417,
|
| 45 |
-
"learning_rate": 1.7346666666666668e-05,
|
| 46 |
-
"loss": 0.09,
|
| 47 |
-
"step": 200
|
| 48 |
-
},
|
| 49 |
-
{
|
| 50 |
-
"epoch": 0.4,
|
| 51 |
-
"eval_accuracy": 0.9895,
|
| 52 |
-
"eval_loss": 0.08769073337316513,
|
| 53 |
-
"eval_runtime": 43.4637,
|
| 54 |
-
"eval_samples_per_second": 46.015,
|
| 55 |
-
"eval_steps_per_second": 5.752,
|
| 56 |
-
"step": 200
|
| 57 |
-
},
|
| 58 |
-
{
|
| 59 |
-
"epoch": 0.5,
|
| 60 |
-
"grad_norm": 0.33664369583129883,
|
| 61 |
-
"learning_rate": 1.668e-05,
|
| 62 |
-
"loss": 0.0704,
|
| 63 |
-
"step": 250
|
| 64 |
-
},
|
| 65 |
-
{
|
| 66 |
-
"epoch": 0.6,
|
| 67 |
-
"grad_norm": 3.125309944152832,
|
| 68 |
-
"learning_rate": 1.6013333333333335e-05,
|
| 69 |
-
"loss": 0.0583,
|
| 70 |
-
"step": 300
|
| 71 |
-
},
|
| 72 |
-
{
|
| 73 |
-
"epoch": 0.6,
|
| 74 |
-
"eval_accuracy": 0.998,
|
| 75 |
-
"eval_loss": 0.04995392635464668,
|
| 76 |
-
"eval_runtime": 42.7949,
|
| 77 |
-
"eval_samples_per_second": 46.735,
|
| 78 |
-
"eval_steps_per_second": 5.842,
|
| 79 |
-
"step": 300
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.7,
|
| 83 |
-
"grad_norm": 0.19215664267539978,
|
| 84 |
-
"learning_rate": 1.5346666666666667e-05,
|
| 85 |
-
"loss": 0.0414,
|
| 86 |
-
"step": 350
|
| 87 |
-
},
|
| 88 |
-
{
|
| 89 |
-
"epoch": 0.8,
|
| 90 |
-
"grad_norm": 0.1444077342748642,
|
| 91 |
-
"learning_rate": 1.4680000000000002e-05,
|
| 92 |
-
"loss": 0.0454,
|
| 93 |
-
"step": 400
|
| 94 |
-
},
|
| 95 |
-
{
|
| 96 |
-
"epoch": 0.8,
|
| 97 |
-
"eval_accuracy": 0.998,
|
| 98 |
-
"eval_loss": 0.03876766189932823,
|
| 99 |
-
"eval_runtime": 42.439,
|
| 100 |
-
"eval_samples_per_second": 47.126,
|
| 101 |
-
"eval_steps_per_second": 5.891,
|
| 102 |
-
"step": 400
|
| 103 |
-
},
|
| 104 |
-
{
|
| 105 |
-
"epoch": 0.9,
|
| 106 |
-
"grad_norm": 0.13351820409297943,
|
| 107 |
-
"learning_rate": 1.4013333333333334e-05,
|
| 108 |
-
"loss": 0.0351,
|
| 109 |
-
"step": 450
|
| 110 |
-
},
|
| 111 |
-
{
|
| 112 |
-
"epoch": 1.0,
|
| 113 |
-
"grad_norm": 1.3633350133895874,
|
| 114 |
-
"learning_rate": 1.3346666666666667e-05,
|
| 115 |
-
"loss": 0.029,
|
| 116 |
-
"step": 500
|
| 117 |
-
},
|
| 118 |
-
{
|
| 119 |
-
"epoch": 1.0,
|
| 120 |
-
"eval_accuracy": 0.9985,
|
| 121 |
-
"eval_loss": 0.03332938253879547,
|
| 122 |
-
"eval_runtime": 42.2001,
|
| 123 |
-
"eval_samples_per_second": 47.393,
|
| 124 |
-
"eval_steps_per_second": 5.924,
|
| 125 |
-
"step": 500
|
| 126 |
-
},
|
| 127 |
-
{
|
| 128 |
-
"epoch": 1.1,
|
| 129 |
-
"grad_norm": 0.12323179095983505,
|
| 130 |
-
"learning_rate": 1.268e-05,
|
| 131 |
-
"loss": 0.0249,
|
| 132 |
-
"step": 550
|
| 133 |
-
},
|
| 134 |
-
{
|
| 135 |
-
"epoch": 1.2,
|
| 136 |
-
"grad_norm": 0.12278539687395096,
|
| 137 |
-
"learning_rate": 1.2013333333333334e-05,
|
| 138 |
-
"loss": 0.0301,
|
| 139 |
-
"step": 600
|
| 140 |
-
},
|
| 141 |
-
{
|
| 142 |
-
"epoch": 1.2,
|
| 143 |
-
"eval_accuracy": 0.9975,
|
| 144 |
-
"eval_loss": 0.031327877193689346,
|
| 145 |
-
"eval_runtime": 42.3539,
|
| 146 |
-
"eval_samples_per_second": 47.221,
|
| 147 |
-
"eval_steps_per_second": 5.903,
|
| 148 |
-
"step": 600
|
| 149 |
-
},
|
| 150 |
-
{
|
| 151 |
-
"epoch": 1.3,
|
| 152 |
-
"grad_norm": 0.12111877650022507,
|
| 153 |
-
"learning_rate": 1.1346666666666668e-05,
|
| 154 |
-
"loss": 0.0322,
|
| 155 |
-
"step": 650
|
| 156 |
-
},
|
| 157 |
-
{
|
| 158 |
-
"epoch": 1.4,
|
| 159 |
-
"grad_norm": 0.10371693968772888,
|
| 160 |
-
"learning_rate": 1.0680000000000001e-05,
|
| 161 |
-
"loss": 0.0243,
|
| 162 |
-
"step": 700
|
| 163 |
-
},
|
| 164 |
-
{
|
| 165 |
-
"epoch": 1.4,
|
| 166 |
-
"eval_accuracy": 0.997,
|
| 167 |
-
"eval_loss": 0.029479531571269035,
|
| 168 |
-
"eval_runtime": 42.2575,
|
| 169 |
-
"eval_samples_per_second": 47.329,
|
| 170 |
-
"eval_steps_per_second": 5.916,
|
| 171 |
-
"step": 700
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 1.5,
|
| 175 |
-
"grad_norm": 0.09562443196773529,
|
| 176 |
-
"learning_rate": 1.0013333333333335e-05,
|
| 177 |
-
"loss": 0.026,
|
| 178 |
-
"step": 750
|
| 179 |
-
},
|
| 180 |
-
{
|
| 181 |
-
"epoch": 1.6,
|
| 182 |
-
"grad_norm": 0.09563508629798889,
|
| 183 |
-
"learning_rate": 9.346666666666666e-06,
|
| 184 |
-
"loss": 0.0181,
|
| 185 |
-
"step": 800
|
| 186 |
-
},
|
| 187 |
-
{
|
| 188 |
-
"epoch": 1.6,
|
| 189 |
-
"eval_accuracy": 0.997,
|
| 190 |
-
"eval_loss": 0.027446668595075607,
|
| 191 |
-
"eval_runtime": 42.1658,
|
| 192 |
-
"eval_samples_per_second": 47.432,
|
| 193 |
-
"eval_steps_per_second": 5.929,
|
| 194 |
-
"step": 800
|
| 195 |
-
},
|
| 196 |
-
{
|
| 197 |
-
"epoch": 1.7,
|
| 198 |
-
"grad_norm": 0.08600907772779465,
|
| 199 |
-
"learning_rate": 8.68e-06,
|
| 200 |
-
"loss": 0.0252,
|
| 201 |
-
"step": 850
|
| 202 |
-
},
|
| 203 |
-
{
|
| 204 |
-
"epoch": 1.8,
|
| 205 |
-
"grad_norm": 0.10505673289299011,
|
| 206 |
-
"learning_rate": 8.013333333333333e-06,
|
| 207 |
-
"loss": 0.0167,
|
| 208 |
-
"step": 900
|
| 209 |
-
},
|
| 210 |
-
{
|
| 211 |
-
"epoch": 1.8,
|
| 212 |
-
"eval_accuracy": 0.997,
|
| 213 |
-
"eval_loss": 0.02535245567560196,
|
| 214 |
-
"eval_runtime": 42.2891,
|
| 215 |
-
"eval_samples_per_second": 47.294,
|
| 216 |
-
"eval_steps_per_second": 5.912,
|
| 217 |
-
"step": 900
|
| 218 |
-
},
|
| 219 |
-
{
|
| 220 |
-
"epoch": 1.9,
|
| 221 |
-
"grad_norm": 0.08181212097406387,
|
| 222 |
-
"learning_rate": 7.346666666666668e-06,
|
| 223 |
-
"loss": 0.016,
|
| 224 |
-
"step": 950
|
| 225 |
-
},
|
| 226 |
-
{
|
| 227 |
-
"epoch": 2.0,
|
| 228 |
-
"grad_norm": 0.07987715303897858,
|
| 229 |
-
"learning_rate": 6.680000000000001e-06,
|
| 230 |
-
"loss": 0.0159,
|
| 231 |
-
"step": 1000
|
| 232 |
-
},
|
| 233 |
-
{
|
| 234 |
-
"epoch": 2.0,
|
| 235 |
-
"eval_accuracy": 0.9955,
|
| 236 |
-
"eval_loss": 0.028334029018878937,
|
| 237 |
-
"eval_runtime": 42.9091,
|
| 238 |
-
"eval_samples_per_second": 46.61,
|
| 239 |
-
"eval_steps_per_second": 5.826,
|
| 240 |
-
"step": 1000
|
| 241 |
-
},
|
| 242 |
-
{
|
| 243 |
-
"epoch": 2.1,
|
| 244 |
-
"grad_norm": 17.688356399536133,
|
| 245 |
-
"learning_rate": 6.013333333333335e-06,
|
| 246 |
-
"loss": 0.0191,
|
| 247 |
-
"step": 1050
|
| 248 |
-
},
|
| 249 |
-
{
|
| 250 |
-
"epoch": 2.2,
|
| 251 |
-
"grad_norm": 0.07392571121454239,
|
| 252 |
-
"learning_rate": 5.346666666666667e-06,
|
| 253 |
-
"loss": 0.0154,
|
| 254 |
-
"step": 1100
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 2.2,
|
| 258 |
-
"eval_accuracy": 0.9985,
|
| 259 |
-
"eval_loss": 0.021651165559887886,
|
| 260 |
-
"eval_runtime": 42.7586,
|
| 261 |
-
"eval_samples_per_second": 46.774,
|
| 262 |
-
"eval_steps_per_second": 5.847,
|
| 263 |
-
"step": 1100
|
| 264 |
-
},
|
| 265 |
-
{
|
| 266 |
-
"epoch": 2.3,
|
| 267 |
-
"grad_norm": 0.09023883938789368,
|
| 268 |
-
"learning_rate": 4.680000000000001e-06,
|
| 269 |
-
"loss": 0.014,
|
| 270 |
-
"step": 1150
|
| 271 |
-
},
|
| 272 |
-
{
|
| 273 |
-
"epoch": 2.4,
|
| 274 |
-
"grad_norm": 0.075103759765625,
|
| 275 |
-
"learning_rate": 4.013333333333334e-06,
|
| 276 |
-
"loss": 0.0155,
|
| 277 |
-
"step": 1200
|
| 278 |
-
},
|
| 279 |
-
{
|
| 280 |
-
"epoch": 2.4,
|
| 281 |
-
"eval_accuracy": 0.998,
|
| 282 |
-
"eval_loss": 0.021380621939897537,
|
| 283 |
-
"eval_runtime": 42.7122,
|
| 284 |
-
"eval_samples_per_second": 46.825,
|
| 285 |
-
"eval_steps_per_second": 5.853,
|
| 286 |
-
"step": 1200
|
| 287 |
-
},
|
| 288 |
-
{
|
| 289 |
-
"epoch": 2.5,
|
| 290 |
-
"grad_norm": 0.07060286402702332,
|
| 291 |
-
"learning_rate": 3.346666666666667e-06,
|
| 292 |
-
"loss": 0.0135,
|
| 293 |
-
"step": 1250
|
| 294 |
-
},
|
| 295 |
-
{
|
| 296 |
-
"epoch": 2.6,
|
| 297 |
-
"grad_norm": 0.06728706508874893,
|
| 298 |
-
"learning_rate": 2.68e-06,
|
| 299 |
-
"loss": 0.0133,
|
| 300 |
-
"step": 1300
|
| 301 |
-
},
|
| 302 |
-
{
|
| 303 |
-
"epoch": 2.6,
|
| 304 |
-
"eval_accuracy": 0.9985,
|
| 305 |
-
"eval_loss": 0.019463825970888138,
|
| 306 |
-
"eval_runtime": 42.4277,
|
| 307 |
-
"eval_samples_per_second": 47.139,
|
| 308 |
-
"eval_steps_per_second": 5.892,
|
| 309 |
-
"step": 1300
|
| 310 |
-
},
|
| 311 |
-
{
|
| 312 |
-
"epoch": 2.7,
|
| 313 |
-
"grad_norm": 0.07170647382736206,
|
| 314 |
-
"learning_rate": 2.0133333333333337e-06,
|
| 315 |
-
"loss": 0.0139,
|
| 316 |
-
"step": 1350
|
| 317 |
-
},
|
| 318 |
-
{
|
| 319 |
-
"epoch": 2.8,
|
| 320 |
-
"grad_norm": 0.07239634543657303,
|
| 321 |
-
"learning_rate": 1.3466666666666668e-06,
|
| 322 |
-
"loss": 0.0129,
|
| 323 |
-
"step": 1400
|
| 324 |
-
},
|
| 325 |
-
{
|
| 326 |
-
"epoch": 2.8,
|
| 327 |
-
"eval_accuracy": 0.9985,
|
| 328 |
-
"eval_loss": 0.019522910937666893,
|
| 329 |
-
"eval_runtime": 42.425,
|
| 330 |
-
"eval_samples_per_second": 47.142,
|
| 331 |
-
"eval_steps_per_second": 5.893,
|
| 332 |
-
"step": 1400
|
| 333 |
-
},
|
| 334 |
-
{
|
| 335 |
-
"epoch": 2.9,
|
| 336 |
-
"grad_norm": 0.07264234870672226,
|
| 337 |
-
"learning_rate": 6.800000000000001e-07,
|
| 338 |
-
"loss": 0.0129,
|
| 339 |
-
"step": 1450
|
| 340 |
-
},
|
| 341 |
-
{
|
| 342 |
-
"epoch": 3.0,
|
| 343 |
-
"grad_norm": 0.07093141227960587,
|
| 344 |
-
"learning_rate": 1.3333333333333334e-08,
|
| 345 |
-
"loss": 0.0128,
|
| 346 |
-
"step": 1500
|
| 347 |
-
},
|
| 348 |
-
{
|
| 349 |
-
"epoch": 3.0,
|
| 350 |
-
"eval_accuracy": 0.9985,
|
| 351 |
-
"eval_loss": 0.0190621018409729,
|
| 352 |
-
"eval_runtime": 42.0295,
|
| 353 |
-
"eval_samples_per_second": 47.586,
|
| 354 |
-
"eval_steps_per_second": 5.948,
|
| 355 |
-
"step": 1500
|
| 356 |
-
}
|
| 357 |
-
],
|
| 358 |
-
"logging_steps": 50,
|
| 359 |
-
"max_steps": 1500,
|
| 360 |
-
"num_input_tokens_seen": 0,
|
| 361 |
-
"num_train_epochs": 3,
|
| 362 |
-
"save_steps": 200,
|
| 363 |
-
"stateful_callbacks": {
|
| 364 |
-
"TrainerControl": {
|
| 365 |
-
"args": {
|
| 366 |
-
"should_epoch_stop": false,
|
| 367 |
-
"should_evaluate": false,
|
| 368 |
-
"should_log": false,
|
| 369 |
-
"should_save": true,
|
| 370 |
-
"should_training_stop": true
|
| 371 |
-
},
|
| 372 |
-
"attributes": {}
|
| 373 |
-
}
|
| 374 |
-
},
|
| 375 |
-
"total_flos": 1.8595919415663636e+18,
|
| 376 |
-
"train_batch_size": 16,
|
| 377 |
-
"trial_name": null,
|
| 378 |
-
"trial_params": null
|
| 379 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/checkpoint-1500/training_args.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:c754b2f54e126b986b7308e6cfafe7689841cb35bd670517252cc89c62c69f1e
|
| 3 |
-
size 5304
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/config.json
DELETED
|
@@ -1,35 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"ViTForImageClassification"
|
| 4 |
-
],
|
| 5 |
-
"attention_probs_dropout_prob": 0.0,
|
| 6 |
-
"encoder_stride": 16,
|
| 7 |
-
"hidden_act": "gelu",
|
| 8 |
-
"hidden_dropout_prob": 0.0,
|
| 9 |
-
"hidden_size": 768,
|
| 10 |
-
"id2label": {
|
| 11 |
-
"0": "AI Generated",
|
| 12 |
-
"1": "Deepfake",
|
| 13 |
-
"2": "Real"
|
| 14 |
-
},
|
| 15 |
-
"image_size": 224,
|
| 16 |
-
"initializer_range": 0.02,
|
| 17 |
-
"intermediate_size": 3072,
|
| 18 |
-
"label2id": {
|
| 19 |
-
"AI Generated": 0,
|
| 20 |
-
"Deepfake": 1,
|
| 21 |
-
"Real": 2
|
| 22 |
-
},
|
| 23 |
-
"layer_norm_eps": 1e-12,
|
| 24 |
-
"model_type": "vit",
|
| 25 |
-
"num_attention_heads": 12,
|
| 26 |
-
"num_channels": 3,
|
| 27 |
-
"num_hidden_layers": 12,
|
| 28 |
-
"patch_size": 16,
|
| 29 |
-
"pooler_act": "tanh",
|
| 30 |
-
"pooler_output_size": 768,
|
| 31 |
-
"problem_type": "single_label_classification",
|
| 32 |
-
"qkv_bias": true,
|
| 33 |
-
"torch_dtype": "float32",
|
| 34 |
-
"transformers_version": "4.53.1"
|
| 35 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/confusion_matrix.png
DELETED
|
Binary file (48.3 kB)
|
|
|
vit-deepfake-model/model.safetensors
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:eee4f6e7a8f62cfa2027345fbe75408298bc7a2f9db2dca567b018e7f0434787
|
| 3 |
-
size 343227052
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/preprocessor_config.json
DELETED
|
@@ -1,23 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"do_convert_rgb": null,
|
| 3 |
-
"do_normalize": true,
|
| 4 |
-
"do_rescale": true,
|
| 5 |
-
"do_resize": true,
|
| 6 |
-
"image_mean": [
|
| 7 |
-
0.5,
|
| 8 |
-
0.5,
|
| 9 |
-
0.5
|
| 10 |
-
],
|
| 11 |
-
"image_processor_type": "ViTImageProcessor",
|
| 12 |
-
"image_std": [
|
| 13 |
-
0.5,
|
| 14 |
-
0.5,
|
| 15 |
-
0.5
|
| 16 |
-
],
|
| 17 |
-
"resample": 2,
|
| 18 |
-
"rescale_factor": 0.00392156862745098,
|
| 19 |
-
"size": {
|
| 20 |
-
"height": 224,
|
| 21 |
-
"width": 224
|
| 22 |
-
}
|
| 23 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
vit-deepfake-model/training_args.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:c754b2f54e126b986b7308e6cfafe7689841cb35bd670517252cc89c62c69f1e
|
| 3 |
-
size 5304
|
|
|
|
|
|
|
|
|
|
|
|