Commit
•
ff577e2
1
Parent(s):
4ef0fd3
Upload trainer_state.json
Browse files- trainer_state.json +370 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,370 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.08029154688119888,
|
| 3 |
+
"best_model_checkpoint": "./beit-tiny-mask-finetuned/checkpoint-200",
|
| 4 |
+
"epoch": 0.5543237250554324,
|
| 5 |
+
"global_step": 500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 0.00019955654101995565,
|
| 13 |
+
"loss": 0.1318,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 0.00019900221729490022,
|
| 19 |
+
"loss": 0.0546,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 0.00019850332594235036,
|
| 25 |
+
"loss": 0.2395,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.04,
|
| 30 |
+
"learning_rate": 0.00019794900221729492,
|
| 31 |
+
"loss": 0.0966,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.06,
|
| 36 |
+
"learning_rate": 0.00019739467849223948,
|
| 37 |
+
"loss": 0.1544,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.07,
|
| 42 |
+
"learning_rate": 0.00019684035476718405,
|
| 43 |
+
"loss": 0.1734,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.08,
|
| 48 |
+
"learning_rate": 0.00019634146341463416,
|
| 49 |
+
"loss": 0.1556,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.09,
|
| 54 |
+
"learning_rate": 0.00019578713968957873,
|
| 55 |
+
"loss": 0.2154,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.1,
|
| 60 |
+
"learning_rate": 0.0001952328159645233,
|
| 61 |
+
"loss": 0.1845,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.11,
|
| 66 |
+
"learning_rate": 0.0001947339246119734,
|
| 67 |
+
"loss": 0.4978,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.11,
|
| 72 |
+
"eval_accuracy": 0.9402113459399333,
|
| 73 |
+
"eval_loss": 0.17701464891433716,
|
| 74 |
+
"eval_runtime": 192.8644,
|
| 75 |
+
"eval_samples_per_second": 18.645,
|
| 76 |
+
"eval_steps_per_second": 2.333,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.12,
|
| 81 |
+
"learning_rate": 0.00019417960088691797,
|
| 82 |
+
"loss": 0.1897,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.13,
|
| 87 |
+
"learning_rate": 0.00019362527716186253,
|
| 88 |
+
"loss": 0.1148,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.14,
|
| 93 |
+
"learning_rate": 0.0001930709534368071,
|
| 94 |
+
"loss": 0.208,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.16,
|
| 99 |
+
"learning_rate": 0.00019251662971175165,
|
| 100 |
+
"loss": 0.1142,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.17,
|
| 105 |
+
"learning_rate": 0.00019196230598669624,
|
| 106 |
+
"loss": 0.1285,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.18,
|
| 111 |
+
"learning_rate": 0.0001914079822616408,
|
| 112 |
+
"loss": 0.1383,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.19,
|
| 117 |
+
"learning_rate": 0.00019085365853658537,
|
| 118 |
+
"loss": 0.1786,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.2,
|
| 123 |
+
"learning_rate": 0.00019029933481152996,
|
| 124 |
+
"loss": 0.0892,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.21,
|
| 129 |
+
"learning_rate": 0.00018974501108647452,
|
| 130 |
+
"loss": 0.2118,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.22,
|
| 135 |
+
"learning_rate": 0.00018919068736141908,
|
| 136 |
+
"loss": 0.1863,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.22,
|
| 141 |
+
"eval_accuracy": 0.974972191323693,
|
| 142 |
+
"eval_loss": 0.08029154688119888,
|
| 143 |
+
"eval_runtime": 193.2131,
|
| 144 |
+
"eval_samples_per_second": 18.612,
|
| 145 |
+
"eval_steps_per_second": 2.329,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.23,
|
| 150 |
+
"learning_rate": 0.00018863636363636364,
|
| 151 |
+
"loss": 0.0495,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.24,
|
| 156 |
+
"learning_rate": 0.0001880820399113082,
|
| 157 |
+
"loss": 0.1476,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.25,
|
| 162 |
+
"learning_rate": 0.00018752771618625277,
|
| 163 |
+
"loss": 0.1339,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.27,
|
| 168 |
+
"learning_rate": 0.00018697339246119733,
|
| 169 |
+
"loss": 0.1734,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.28,
|
| 174 |
+
"learning_rate": 0.00018641906873614192,
|
| 175 |
+
"loss": 0.2068,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.29,
|
| 180 |
+
"learning_rate": 0.00018586474501108648,
|
| 181 |
+
"loss": 0.1427,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.3,
|
| 186 |
+
"learning_rate": 0.00018531042128603104,
|
| 187 |
+
"loss": 0.182,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.31,
|
| 192 |
+
"learning_rate": 0.00018475609756097563,
|
| 193 |
+
"loss": 0.1436,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.32,
|
| 198 |
+
"learning_rate": 0.0001842017738359202,
|
| 199 |
+
"loss": 0.1319,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.33,
|
| 204 |
+
"learning_rate": 0.00018364745011086476,
|
| 205 |
+
"loss": 0.0818,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.33,
|
| 210 |
+
"eval_accuracy": 0.9424360400444939,
|
| 211 |
+
"eval_loss": 0.20646639168262482,
|
| 212 |
+
"eval_runtime": 193.0709,
|
| 213 |
+
"eval_samples_per_second": 18.625,
|
| 214 |
+
"eval_steps_per_second": 2.331,
|
| 215 |
+
"step": 300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.34,
|
| 219 |
+
"learning_rate": 0.00018309312638580932,
|
| 220 |
+
"loss": 0.2115,
|
| 221 |
+
"step": 310
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.35,
|
| 225 |
+
"learning_rate": 0.00018253880266075388,
|
| 226 |
+
"loss": 0.0725,
|
| 227 |
+
"step": 320
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.37,
|
| 231 |
+
"learning_rate": 0.00018198447893569844,
|
| 232 |
+
"loss": 0.0589,
|
| 233 |
+
"step": 330
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.38,
|
| 237 |
+
"learning_rate": 0.000181430155210643,
|
| 238 |
+
"loss": 0.2017,
|
| 239 |
+
"step": 340
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 0.39,
|
| 243 |
+
"learning_rate": 0.0001808758314855876,
|
| 244 |
+
"loss": 0.0678,
|
| 245 |
+
"step": 350
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 0.4,
|
| 249 |
+
"learning_rate": 0.00018032150776053216,
|
| 250 |
+
"loss": 0.2075,
|
| 251 |
+
"step": 360
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.41,
|
| 255 |
+
"learning_rate": 0.00017976718403547672,
|
| 256 |
+
"loss": 0.3123,
|
| 257 |
+
"step": 370
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.42,
|
| 261 |
+
"learning_rate": 0.0001792128603104213,
|
| 262 |
+
"loss": 0.0857,
|
| 263 |
+
"step": 380
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 0.43,
|
| 267 |
+
"learning_rate": 0.00017865853658536587,
|
| 268 |
+
"loss": 0.0958,
|
| 269 |
+
"step": 390
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.44,
|
| 273 |
+
"learning_rate": 0.00017810421286031043,
|
| 274 |
+
"loss": 0.1982,
|
| 275 |
+
"step": 400
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.44,
|
| 279 |
+
"eval_accuracy": 0.9438264738598443,
|
| 280 |
+
"eval_loss": 0.16823017597198486,
|
| 281 |
+
"eval_runtime": 193.1602,
|
| 282 |
+
"eval_samples_per_second": 18.617,
|
| 283 |
+
"eval_steps_per_second": 2.33,
|
| 284 |
+
"step": 400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.45,
|
| 288 |
+
"learning_rate": 0.000177549889135255,
|
| 289 |
+
"loss": 0.2038,
|
| 290 |
+
"step": 410
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.47,
|
| 294 |
+
"learning_rate": 0.00017699556541019956,
|
| 295 |
+
"loss": 0.1603,
|
| 296 |
+
"step": 420
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.48,
|
| 300 |
+
"learning_rate": 0.00017644124168514412,
|
| 301 |
+
"loss": 0.1035,
|
| 302 |
+
"step": 430
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.49,
|
| 306 |
+
"learning_rate": 0.00017588691796008868,
|
| 307 |
+
"loss": 0.0329,
|
| 308 |
+
"step": 440
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.5,
|
| 312 |
+
"learning_rate": 0.00017533259423503327,
|
| 313 |
+
"loss": 0.0392,
|
| 314 |
+
"step": 450
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.51,
|
| 318 |
+
"learning_rate": 0.00017477827050997783,
|
| 319 |
+
"loss": 0.1546,
|
| 320 |
+
"step": 460
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.52,
|
| 324 |
+
"learning_rate": 0.0001742239467849224,
|
| 325 |
+
"loss": 0.155,
|
| 326 |
+
"step": 470
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.53,
|
| 330 |
+
"learning_rate": 0.00017366962305986699,
|
| 331 |
+
"loss": 0.1521,
|
| 332 |
+
"step": 480
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.54,
|
| 336 |
+
"learning_rate": 0.00017311529933481155,
|
| 337 |
+
"loss": 0.1506,
|
| 338 |
+
"step": 490
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.55,
|
| 342 |
+
"learning_rate": 0.0001725609756097561,
|
| 343 |
+
"loss": 0.0293,
|
| 344 |
+
"step": 500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.55,
|
| 348 |
+
"eval_accuracy": 0.9824805339265851,
|
| 349 |
+
"eval_loss": 0.08882919698953629,
|
| 350 |
+
"eval_runtime": 193.0128,
|
| 351 |
+
"eval_samples_per_second": 18.631,
|
| 352 |
+
"eval_steps_per_second": 2.331,
|
| 353 |
+
"step": 500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.55,
|
| 357 |
+
"step": 500,
|
| 358 |
+
"total_flos": 6.18978356527104e+17,
|
| 359 |
+
"train_loss": 0.1509909844994545,
|
| 360 |
+
"train_runtime": 3496.5643,
|
| 361 |
+
"train_samples_per_second": 16.505,
|
| 362 |
+
"train_steps_per_second": 1.032
|
| 363 |
+
}
|
| 364 |
+
],
|
| 365 |
+
"max_steps": 3608,
|
| 366 |
+
"num_train_epochs": 4,
|
| 367 |
+
"total_flos": 6.18978356527104e+17,
|
| 368 |
+
"trial_name": null,
|
| 369 |
+
"trial_params": null
|
| 370 |
+
}
|