Commit
•
dd85623
1
Parent(s):
c618a1d
Training in progress, epoch 0
Browse files- all_results.json +8 -0
- pytorch_model.bin +1 -1
- test_results.json +8 -0
- trainer_state.json +313 -0
- training_args.bin +1 -1
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.24,
|
| 3 |
+
"eval_accuracy": 0.8138686131386861,
|
| 4 |
+
"eval_loss": 0.6085803508758545,
|
| 5 |
+
"eval_runtime": 2468.738,
|
| 6 |
+
"eval_samples_per_second": 0.444,
|
| 7 |
+
"eval_steps_per_second": 0.055
|
| 8 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 345031669
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6911e3117ec5edd75e6e6ee570b211aac37c69509a0a4d60242e2a058de0636f
|
| 3 |
size 345031669
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.24,
|
| 3 |
+
"eval_accuracy": 0.8138686131386861,
|
| 4 |
+
"eval_loss": 0.6085803508758545,
|
| 5 |
+
"eval_runtime": 2468.738,
|
| 6 |
+
"eval_samples_per_second": 0.444,
|
| 7 |
+
"eval_steps_per_second": 0.055
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,313 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.804093567251462,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ssv2-finetuned-samsung-subset/checkpoint-408",
|
| 4 |
+
"epoch": 3.2426470588235294,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 408,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 1.2195121951219513e-05,
|
| 14 |
+
"loss": 3.0442,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.05,
|
| 19 |
+
"learning_rate": 2.4390243902439026e-05,
|
| 20 |
+
"loss": 2.9549,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.07,
|
| 25 |
+
"learning_rate": 3.6585365853658535e-05,
|
| 26 |
+
"loss": 2.8135,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.1,
|
| 31 |
+
"learning_rate": 4.878048780487805e-05,
|
| 32 |
+
"loss": 2.6622,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.12,
|
| 37 |
+
"learning_rate": 4.877384196185286e-05,
|
| 38 |
+
"loss": 2.4479,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.15,
|
| 43 |
+
"learning_rate": 4.741144414168938e-05,
|
| 44 |
+
"loss": 2.282,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.17,
|
| 49 |
+
"learning_rate": 4.604904632152589e-05,
|
| 50 |
+
"loss": 1.9786,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2,
|
| 55 |
+
"learning_rate": 4.46866485013624e-05,
|
| 56 |
+
"loss": 1.9135,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.22,
|
| 61 |
+
"learning_rate": 4.332425068119891e-05,
|
| 62 |
+
"loss": 1.6138,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.25,
|
| 67 |
+
"learning_rate": 4.196185286103542e-05,
|
| 68 |
+
"loss": 1.608,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.25,
|
| 73 |
+
"eval_accuracy": 0.5497076023391813,
|
| 74 |
+
"eval_loss": 1.5651301145553589,
|
| 75 |
+
"eval_runtime": 3380.9023,
|
| 76 |
+
"eval_samples_per_second": 0.303,
|
| 77 |
+
"eval_steps_per_second": 0.038,
|
| 78 |
+
"step": 103
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.02,
|
| 82 |
+
"learning_rate": 4.0599455040871935e-05,
|
| 83 |
+
"loss": 1.3634,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.04,
|
| 88 |
+
"learning_rate": 3.923705722070845e-05,
|
| 89 |
+
"loss": 1.3086,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.07,
|
| 94 |
+
"learning_rate": 3.787465940054496e-05,
|
| 95 |
+
"loss": 1.1414,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.09,
|
| 100 |
+
"learning_rate": 3.651226158038147e-05,
|
| 101 |
+
"loss": 1.0196,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.12,
|
| 106 |
+
"learning_rate": 3.5149863760217985e-05,
|
| 107 |
+
"loss": 0.9671,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.14,
|
| 112 |
+
"learning_rate": 3.37874659400545e-05,
|
| 113 |
+
"loss": 0.9137,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.16,
|
| 118 |
+
"learning_rate": 3.242506811989101e-05,
|
| 119 |
+
"loss": 0.9068,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.19,
|
| 124 |
+
"learning_rate": 3.106267029972752e-05,
|
| 125 |
+
"loss": 1.0022,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.21,
|
| 130 |
+
"learning_rate": 2.9700272479564035e-05,
|
| 131 |
+
"loss": 0.8898,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.24,
|
| 136 |
+
"learning_rate": 2.8337874659400547e-05,
|
| 137 |
+
"loss": 0.7421,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.25,
|
| 142 |
+
"eval_accuracy": 0.7270955165692008,
|
| 143 |
+
"eval_loss": 0.9049609303474426,
|
| 144 |
+
"eval_runtime": 3202.4383,
|
| 145 |
+
"eval_samples_per_second": 0.32,
|
| 146 |
+
"eval_steps_per_second": 0.04,
|
| 147 |
+
"step": 206
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.01,
|
| 151 |
+
"learning_rate": 2.6975476839237056e-05,
|
| 152 |
+
"loss": 0.7335,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.03,
|
| 157 |
+
"learning_rate": 2.5613079019073572e-05,
|
| 158 |
+
"loss": 0.7486,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.06,
|
| 163 |
+
"learning_rate": 2.425068119891008e-05,
|
| 164 |
+
"loss": 0.6743,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.08,
|
| 169 |
+
"learning_rate": 2.2888283378746594e-05,
|
| 170 |
+
"loss": 0.5636,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.11,
|
| 175 |
+
"learning_rate": 2.1525885558583106e-05,
|
| 176 |
+
"loss": 0.6737,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.13,
|
| 181 |
+
"learning_rate": 2.016348773841962e-05,
|
| 182 |
+
"loss": 0.465,
|
| 183 |
+
"step": 260
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.16,
|
| 187 |
+
"learning_rate": 1.880108991825613e-05,
|
| 188 |
+
"loss": 0.6182,
|
| 189 |
+
"step": 270
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.18,
|
| 193 |
+
"learning_rate": 1.7438692098092644e-05,
|
| 194 |
+
"loss": 0.5307,
|
| 195 |
+
"step": 280
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.21,
|
| 199 |
+
"learning_rate": 1.6076294277929156e-05,
|
| 200 |
+
"loss": 0.503,
|
| 201 |
+
"step": 290
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.23,
|
| 205 |
+
"learning_rate": 1.4713896457765669e-05,
|
| 206 |
+
"loss": 0.6384,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.25,
|
| 211 |
+
"eval_accuracy": 0.7962962962962963,
|
| 212 |
+
"eval_loss": 0.6470026969909668,
|
| 213 |
+
"eval_runtime": 3136.7537,
|
| 214 |
+
"eval_samples_per_second": 0.327,
|
| 215 |
+
"eval_steps_per_second": 0.041,
|
| 216 |
+
"step": 309
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.0,
|
| 220 |
+
"learning_rate": 1.3351498637602181e-05,
|
| 221 |
+
"loss": 0.4937,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.03,
|
| 226 |
+
"learning_rate": 1.1989100817438692e-05,
|
| 227 |
+
"loss": 0.3847,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.05,
|
| 232 |
+
"learning_rate": 1.0626702997275204e-05,
|
| 233 |
+
"loss": 0.3882,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 3.08,
|
| 238 |
+
"learning_rate": 9.264305177111717e-06,
|
| 239 |
+
"loss": 0.45,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.1,
|
| 244 |
+
"learning_rate": 7.90190735694823e-06,
|
| 245 |
+
"loss": 0.3976,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.12,
|
| 250 |
+
"learning_rate": 6.539509536784741e-06,
|
| 251 |
+
"loss": 0.4559,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 3.15,
|
| 256 |
+
"learning_rate": 5.1771117166212534e-06,
|
| 257 |
+
"loss": 0.4666,
|
| 258 |
+
"step": 370
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 3.17,
|
| 262 |
+
"learning_rate": 3.814713896457766e-06,
|
| 263 |
+
"loss": 0.38,
|
| 264 |
+
"step": 380
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.2,
|
| 268 |
+
"learning_rate": 2.452316076294278e-06,
|
| 269 |
+
"loss": 0.3852,
|
| 270 |
+
"step": 390
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 3.22,
|
| 274 |
+
"learning_rate": 1.0899182561307902e-06,
|
| 275 |
+
"loss": 0.3902,
|
| 276 |
+
"step": 400
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 3.24,
|
| 280 |
+
"eval_accuracy": 0.804093567251462,
|
| 281 |
+
"eval_loss": 0.5762231945991516,
|
| 282 |
+
"eval_runtime": 2421.5515,
|
| 283 |
+
"eval_samples_per_second": 0.424,
|
| 284 |
+
"eval_steps_per_second": 0.053,
|
| 285 |
+
"step": 408
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 3.24,
|
| 289 |
+
"step": 408,
|
| 290 |
+
"total_flos": 4.056591324738355e+18,
|
| 291 |
+
"train_loss": 1.0837838474441976,
|
| 292 |
+
"train_runtime": 27835.4406,
|
| 293 |
+
"train_samples_per_second": 0.117,
|
| 294 |
+
"train_steps_per_second": 0.015
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 3.24,
|
| 298 |
+
"eval_accuracy": 0.8138686131386861,
|
| 299 |
+
"eval_loss": 0.6085803508758545,
|
| 300 |
+
"eval_runtime": 2468.738,
|
| 301 |
+
"eval_samples_per_second": 0.444,
|
| 302 |
+
"eval_steps_per_second": 0.055,
|
| 303 |
+
"step": 408
|
| 304 |
+
}
|
| 305 |
+
],
|
| 306 |
+
"logging_steps": 10,
|
| 307 |
+
"max_steps": 408,
|
| 308 |
+
"num_train_epochs": 9223372036854775807,
|
| 309 |
+
"save_steps": 500,
|
| 310 |
+
"total_flos": 4.056591324738355e+18,
|
| 311 |
+
"trial_name": null,
|
| 312 |
+
"trial_params": null
|
| 313 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4155
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e30127f7052998e706e6a0f0c48e6c03d05ffdec964a6f52056b2b83679afda3
|
| 3 |
size 4155
|