Commit
•
a16eb88
1
Parent(s):
ea9f00f
End of training
Browse files- trainer_state.json +305 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,305 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.5942937135696411,
|
| 3 |
+
"best_model_checkpoint": "detr-r101-cd45rb-8ah-6l-256d-4096ffn-correcetd\\checkpoint-92120",
|
| 4 |
+
"epoch": 20.0,
|
| 5 |
+
"global_step": 92120,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"learning_rate": 9.501085540599219e-06,
|
| 13 |
+
"loss": 2.3194,
|
| 14 |
+
"step": 4606
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_loss": 1.7874349355697632,
|
| 19 |
+
"eval_runtime": 255.9695,
|
| 20 |
+
"eval_samples_per_second": 6.958,
|
| 21 |
+
"eval_steps_per_second": 0.871,
|
| 22 |
+
"step": 4606
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 2.0,
|
| 26 |
+
"learning_rate": 9.001085540599219e-06,
|
| 27 |
+
"loss": 2.3461,
|
| 28 |
+
"step": 9212
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"eval_loss": 1.8375755548477173,
|
| 33 |
+
"eval_runtime": 255.8256,
|
| 34 |
+
"eval_samples_per_second": 6.962,
|
| 35 |
+
"eval_steps_per_second": 0.872,
|
| 36 |
+
"step": 9212
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 3.0,
|
| 40 |
+
"learning_rate": 8.501302648719063e-06,
|
| 41 |
+
"loss": 2.3028,
|
| 42 |
+
"step": 13818
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.0,
|
| 46 |
+
"eval_loss": 1.9087122678756714,
|
| 47 |
+
"eval_runtime": 255.4494,
|
| 48 |
+
"eval_samples_per_second": 6.972,
|
| 49 |
+
"eval_steps_per_second": 0.873,
|
| 50 |
+
"step": 13818
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"learning_rate": 8.001628310898828e-06,
|
| 55 |
+
"loss": 2.2979,
|
| 56 |
+
"step": 18424
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.0,
|
| 60 |
+
"eval_loss": 1.906615972518921,
|
| 61 |
+
"eval_runtime": 255.8122,
|
| 62 |
+
"eval_samples_per_second": 6.962,
|
| 63 |
+
"eval_steps_per_second": 0.872,
|
| 64 |
+
"step": 18424
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 5.0,
|
| 68 |
+
"learning_rate": 7.501736864958749e-06,
|
| 69 |
+
"loss": 2.273,
|
| 70 |
+
"step": 23030
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 5.0,
|
| 74 |
+
"eval_loss": 1.7688573598861694,
|
| 75 |
+
"eval_runtime": 223.3286,
|
| 76 |
+
"eval_samples_per_second": 7.975,
|
| 77 |
+
"eval_steps_per_second": 0.999,
|
| 78 |
+
"step": 23030
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 6.0,
|
| 82 |
+
"learning_rate": 7.001845419018671e-06,
|
| 83 |
+
"loss": 2.2372,
|
| 84 |
+
"step": 27636
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.0,
|
| 88 |
+
"eval_loss": 1.850079894065857,
|
| 89 |
+
"eval_runtime": 222.161,
|
| 90 |
+
"eval_samples_per_second": 8.017,
|
| 91 |
+
"eval_steps_per_second": 1.004,
|
| 92 |
+
"step": 27636
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 7.0,
|
| 96 |
+
"learning_rate": 6.501953973078594e-06,
|
| 97 |
+
"loss": 2.2429,
|
| 98 |
+
"step": 32242
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.0,
|
| 102 |
+
"eval_loss": 1.7625480890274048,
|
| 103 |
+
"eval_runtime": 221.7461,
|
| 104 |
+
"eval_samples_per_second": 8.032,
|
| 105 |
+
"eval_steps_per_second": 1.006,
|
| 106 |
+
"step": 32242
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 8.0,
|
| 110 |
+
"learning_rate": 6.0021710811984364e-06,
|
| 111 |
+
"loss": 2.2066,
|
| 112 |
+
"step": 36848
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 8.0,
|
| 116 |
+
"eval_loss": 1.7588337659835815,
|
| 117 |
+
"eval_runtime": 221.9488,
|
| 118 |
+
"eval_samples_per_second": 8.024,
|
| 119 |
+
"eval_steps_per_second": 1.005,
|
| 120 |
+
"step": 36848
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 9.0,
|
| 124 |
+
"learning_rate": 5.5023881893182805e-06,
|
| 125 |
+
"loss": 2.1914,
|
| 126 |
+
"step": 41454
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 9.0,
|
| 130 |
+
"eval_loss": 1.7061983346939087,
|
| 131 |
+
"eval_runtime": 260.572,
|
| 132 |
+
"eval_samples_per_second": 6.835,
|
| 133 |
+
"eval_steps_per_second": 0.856,
|
| 134 |
+
"step": 41454
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.0,
|
| 138 |
+
"learning_rate": 5.002605297438125e-06,
|
| 139 |
+
"loss": 2.1553,
|
| 140 |
+
"step": 46060
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 10.0,
|
| 144 |
+
"eval_loss": 1.706916093826294,
|
| 145 |
+
"eval_runtime": 254.0051,
|
| 146 |
+
"eval_samples_per_second": 7.012,
|
| 147 |
+
"eval_steps_per_second": 0.878,
|
| 148 |
+
"step": 46060
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 11.0,
|
| 152 |
+
"learning_rate": 4.502822405557969e-06,
|
| 153 |
+
"loss": 2.1302,
|
| 154 |
+
"step": 50666
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 11.0,
|
| 158 |
+
"eval_loss": 1.6863188743591309,
|
| 159 |
+
"eval_runtime": 253.207,
|
| 160 |
+
"eval_samples_per_second": 7.034,
|
| 161 |
+
"eval_steps_per_second": 0.881,
|
| 162 |
+
"step": 50666
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 12.0,
|
| 166 |
+
"learning_rate": 4.003039513677812e-06,
|
| 167 |
+
"loss": 2.1321,
|
| 168 |
+
"step": 55272
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 12.0,
|
| 172 |
+
"eval_loss": 1.7721848487854004,
|
| 173 |
+
"eval_runtime": 253.9074,
|
| 174 |
+
"eval_samples_per_second": 7.014,
|
| 175 |
+
"eval_steps_per_second": 0.878,
|
| 176 |
+
"step": 55272
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 13.0,
|
| 180 |
+
"learning_rate": 3.503148067737734e-06,
|
| 181 |
+
"loss": 2.1227,
|
| 182 |
+
"step": 59878
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.0,
|
| 186 |
+
"eval_loss": 1.6639100313186646,
|
| 187 |
+
"eval_runtime": 254.2906,
|
| 188 |
+
"eval_samples_per_second": 7.004,
|
| 189 |
+
"eval_steps_per_second": 0.877,
|
| 190 |
+
"step": 59878
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 14.0,
|
| 194 |
+
"learning_rate": 3.003365175857577e-06,
|
| 195 |
+
"loss": 2.087,
|
| 196 |
+
"step": 64484
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 14.0,
|
| 200 |
+
"eval_loss": 1.6486269235610962,
|
| 201 |
+
"eval_runtime": 254.2198,
|
| 202 |
+
"eval_samples_per_second": 7.006,
|
| 203 |
+
"eval_steps_per_second": 0.877,
|
| 204 |
+
"step": 64484
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 15.0,
|
| 208 |
+
"learning_rate": 2.5036908380373426e-06,
|
| 209 |
+
"loss": 2.0676,
|
| 210 |
+
"step": 69090
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 15.0,
|
| 214 |
+
"eval_loss": 1.6632195711135864,
|
| 215 |
+
"eval_runtime": 253.8344,
|
| 216 |
+
"eval_samples_per_second": 7.016,
|
| 217 |
+
"eval_steps_per_second": 0.879,
|
| 218 |
+
"step": 69090
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 16.0,
|
| 222 |
+
"learning_rate": 2.0036908380373427e-06,
|
| 223 |
+
"loss": 2.0604,
|
| 224 |
+
"step": 73696
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 16.0,
|
| 228 |
+
"eval_loss": 1.6428617238998413,
|
| 229 |
+
"eval_runtime": 252.8055,
|
| 230 |
+
"eval_samples_per_second": 7.045,
|
| 231 |
+
"eval_steps_per_second": 0.882,
|
| 232 |
+
"step": 73696
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 17.0,
|
| 236 |
+
"learning_rate": 1.5039079461571866e-06,
|
| 237 |
+
"loss": 2.0435,
|
| 238 |
+
"step": 78302
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 17.0,
|
| 242 |
+
"eval_loss": 1.6223891973495483,
|
| 243 |
+
"eval_runtime": 252.2863,
|
| 244 |
+
"eval_samples_per_second": 7.059,
|
| 245 |
+
"eval_steps_per_second": 0.884,
|
| 246 |
+
"step": 78302
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 18.0,
|
| 250 |
+
"learning_rate": 1.004233608336952e-06,
|
| 251 |
+
"loss": 2.0241,
|
| 252 |
+
"step": 82908
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 18.0,
|
| 256 |
+
"eval_loss": 1.632068157196045,
|
| 257 |
+
"eval_runtime": 255.263,
|
| 258 |
+
"eval_samples_per_second": 6.977,
|
| 259 |
+
"eval_steps_per_second": 0.874,
|
| 260 |
+
"step": 82908
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 19.0,
|
| 264 |
+
"learning_rate": 5.044507164567955e-07,
|
| 265 |
+
"loss": 2.0041,
|
| 266 |
+
"step": 87514
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 19.0,
|
| 270 |
+
"eval_loss": 1.5970158576965332,
|
| 271 |
+
"eval_runtime": 252.5106,
|
| 272 |
+
"eval_samples_per_second": 7.053,
|
| 273 |
+
"eval_steps_per_second": 0.883,
|
| 274 |
+
"step": 87514
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 20.0,
|
| 278 |
+
"learning_rate": 4.559270516717325e-09,
|
| 279 |
+
"loss": 1.9966,
|
| 280 |
+
"step": 92120
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 20.0,
|
| 284 |
+
"eval_loss": 1.5942937135696411,
|
| 285 |
+
"eval_runtime": 253.5474,
|
| 286 |
+
"eval_samples_per_second": 7.024,
|
| 287 |
+
"eval_steps_per_second": 0.88,
|
| 288 |
+
"step": 92120
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 20.0,
|
| 292 |
+
"step": 92120,
|
| 293 |
+
"total_flos": 3.09958520212224e+20,
|
| 294 |
+
"train_loss": 2.1620497868778497,
|
| 295 |
+
"train_runtime": 79274.0231,
|
| 296 |
+
"train_samples_per_second": 4.647,
|
| 297 |
+
"train_steps_per_second": 1.162
|
| 298 |
+
}
|
| 299 |
+
],
|
| 300 |
+
"max_steps": 92120,
|
| 301 |
+
"num_train_epochs": 20,
|
| 302 |
+
"total_flos": 3.09958520212224e+20,
|
| 303 |
+
"trial_name": null,
|
| 304 |
+
"trial_params": null
|
| 305 |
+
}
|