Commit
•
03110e0
1
Parent(s):
66f8878
End of training
Browse files- trainer_state.json +375 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,375 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.7369647026062012,
|
| 3 |
+
"best_model_checkpoint": "detr-r50-cd45rb-8ah-6l-500q-corrected\\checkpoint-115150",
|
| 4 |
+
"epoch": 25.0,
|
| 5 |
+
"global_step": 115150,
|
| 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": 9.600868432479375e-06,
|
| 13 |
+
"loss": 1.9439,
|
| 14 |
+
"step": 4606
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_loss": 2.0401558876037598,
|
| 19 |
+
"eval_runtime": 1382.6016,
|
| 20 |
+
"eval_samples_per_second": 1.288,
|
| 21 |
+
"eval_steps_per_second": 0.161,
|
| 22 |
+
"step": 4606
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 2.0,
|
| 26 |
+
"learning_rate": 9.201042118975251e-06,
|
| 27 |
+
"loss": 1.91,
|
| 28 |
+
"step": 9212
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"eval_loss": 1.9975517988204956,
|
| 33 |
+
"eval_runtime": 1419.7704,
|
| 34 |
+
"eval_samples_per_second": 1.254,
|
| 35 |
+
"eval_steps_per_second": 0.157,
|
| 36 |
+
"step": 9212
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 3.0,
|
| 40 |
+
"learning_rate": 8.801302648719062e-06,
|
| 41 |
+
"loss": 1.8797,
|
| 42 |
+
"step": 13818
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.0,
|
| 46 |
+
"eval_loss": 1.9800050258636475,
|
| 47 |
+
"eval_runtime": 1300.0427,
|
| 48 |
+
"eval_samples_per_second": 1.37,
|
| 49 |
+
"eval_steps_per_second": 0.172,
|
| 50 |
+
"step": 13818
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"learning_rate": 8.401476335214937e-06,
|
| 55 |
+
"loss": 1.8557,
|
| 56 |
+
"step": 18424
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.0,
|
| 60 |
+
"eval_loss": 1.9186592102050781,
|
| 61 |
+
"eval_runtime": 1396.4555,
|
| 62 |
+
"eval_samples_per_second": 1.275,
|
| 63 |
+
"eval_steps_per_second": 0.16,
|
| 64 |
+
"step": 18424
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 5.0,
|
| 68 |
+
"learning_rate": 8.001563178462876e-06,
|
| 69 |
+
"loss": 1.8409,
|
| 70 |
+
"step": 23030
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 5.0,
|
| 74 |
+
"eval_loss": 1.9056885242462158,
|
| 75 |
+
"eval_runtime": 707.528,
|
| 76 |
+
"eval_samples_per_second": 2.517,
|
| 77 |
+
"eval_steps_per_second": 0.315,
|
| 78 |
+
"step": 23030
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 6.0,
|
| 82 |
+
"learning_rate": 7.60173686495875e-06,
|
| 83 |
+
"loss": 1.8149,
|
| 84 |
+
"step": 27636
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.0,
|
| 88 |
+
"eval_loss": 1.9150078296661377,
|
| 89 |
+
"eval_runtime": 732.1035,
|
| 90 |
+
"eval_samples_per_second": 2.433,
|
| 91 |
+
"eval_steps_per_second": 0.305,
|
| 92 |
+
"step": 27636
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 7.0,
|
| 96 |
+
"learning_rate": 7.2017368649587506e-06,
|
| 97 |
+
"loss": 1.8095,
|
| 98 |
+
"step": 32242
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.0,
|
| 102 |
+
"eval_loss": 1.8740921020507812,
|
| 103 |
+
"eval_runtime": 740.742,
|
| 104 |
+
"eval_samples_per_second": 2.404,
|
| 105 |
+
"eval_steps_per_second": 0.301,
|
| 106 |
+
"step": 32242
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 8.0,
|
| 110 |
+
"learning_rate": 6.8019105514546245e-06,
|
| 111 |
+
"loss": 1.7931,
|
| 112 |
+
"step": 36848
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 8.0,
|
| 116 |
+
"eval_loss": 1.8915313482284546,
|
| 117 |
+
"eval_runtime": 393.6231,
|
| 118 |
+
"eval_samples_per_second": 4.525,
|
| 119 |
+
"eval_steps_per_second": 0.567,
|
| 120 |
+
"step": 36848
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 9.0,
|
| 124 |
+
"learning_rate": 6.4019973947025625e-06,
|
| 125 |
+
"loss": 1.7758,
|
| 126 |
+
"step": 41454
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 9.0,
|
| 130 |
+
"eval_loss": 1.8704705238342285,
|
| 131 |
+
"eval_runtime": 398.0752,
|
| 132 |
+
"eval_samples_per_second": 4.474,
|
| 133 |
+
"eval_steps_per_second": 0.56,
|
| 134 |
+
"step": 41454
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.0,
|
| 138 |
+
"learning_rate": 6.0020842379505e-06,
|
| 139 |
+
"loss": 1.7674,
|
| 140 |
+
"step": 46060
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 10.0,
|
| 144 |
+
"eval_loss": 1.850295066833496,
|
| 145 |
+
"eval_runtime": 534.0967,
|
| 146 |
+
"eval_samples_per_second": 3.335,
|
| 147 |
+
"eval_steps_per_second": 0.418,
|
| 148 |
+
"step": 46060
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 11.0,
|
| 152 |
+
"learning_rate": 5.602344767694312e-06,
|
| 153 |
+
"loss": 1.7459,
|
| 154 |
+
"step": 50666
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 11.0,
|
| 158 |
+
"eval_loss": 1.818764328956604,
|
| 159 |
+
"eval_runtime": 696.4058,
|
| 160 |
+
"eval_samples_per_second": 2.557,
|
| 161 |
+
"eval_steps_per_second": 0.32,
|
| 162 |
+
"step": 50666
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 12.0,
|
| 166 |
+
"learning_rate": 5.202518454190188e-06,
|
| 167 |
+
"loss": 1.7322,
|
| 168 |
+
"step": 55272
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 12.0,
|
| 172 |
+
"eval_loss": 1.8162474632263184,
|
| 173 |
+
"eval_runtime": 720.9778,
|
| 174 |
+
"eval_samples_per_second": 2.47,
|
| 175 |
+
"eval_steps_per_second": 0.309,
|
| 176 |
+
"step": 55272
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 13.0,
|
| 180 |
+
"learning_rate": 4.8026921406860624e-06,
|
| 181 |
+
"loss": 1.721,
|
| 182 |
+
"step": 59878
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 13.0,
|
| 186 |
+
"eval_loss": 1.800376057624817,
|
| 187 |
+
"eval_runtime": 677.5807,
|
| 188 |
+
"eval_samples_per_second": 2.628,
|
| 189 |
+
"eval_steps_per_second": 0.329,
|
| 190 |
+
"step": 59878
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 14.0,
|
| 194 |
+
"learning_rate": 4.402778983934e-06,
|
| 195 |
+
"loss": 1.7109,
|
| 196 |
+
"step": 64484
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 14.0,
|
| 200 |
+
"eval_loss": 1.7954081296920776,
|
| 201 |
+
"eval_runtime": 630.3996,
|
| 202 |
+
"eval_samples_per_second": 2.825,
|
| 203 |
+
"eval_steps_per_second": 0.354,
|
| 204 |
+
"step": 64484
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 15.0,
|
| 208 |
+
"learning_rate": 4.002952670429874e-06,
|
| 209 |
+
"loss": 1.7027,
|
| 210 |
+
"step": 69090
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 15.0,
|
| 214 |
+
"eval_loss": 1.7816314697265625,
|
| 215 |
+
"eval_runtime": 631.78,
|
| 216 |
+
"eval_samples_per_second": 2.819,
|
| 217 |
+
"eval_steps_per_second": 0.353,
|
| 218 |
+
"step": 69090
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 16.0,
|
| 222 |
+
"learning_rate": 3.603039513677812e-06,
|
| 223 |
+
"loss": 1.6971,
|
| 224 |
+
"step": 73696
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 16.0,
|
| 228 |
+
"eval_loss": 1.7844570875167847,
|
| 229 |
+
"eval_runtime": 632.2022,
|
| 230 |
+
"eval_samples_per_second": 2.817,
|
| 231 |
+
"eval_steps_per_second": 0.353,
|
| 232 |
+
"step": 73696
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 17.0,
|
| 236 |
+
"learning_rate": 3.2033000434216244e-06,
|
| 237 |
+
"loss": 1.6812,
|
| 238 |
+
"step": 78302
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 17.0,
|
| 242 |
+
"eval_loss": 1.7753602266311646,
|
| 243 |
+
"eval_runtime": 625.6408,
|
| 244 |
+
"eval_samples_per_second": 2.847,
|
| 245 |
+
"eval_steps_per_second": 0.356,
|
| 246 |
+
"step": 78302
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 18.0,
|
| 250 |
+
"learning_rate": 2.803300043421624e-06,
|
| 251 |
+
"loss": 1.6735,
|
| 252 |
+
"step": 82908
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 18.0,
|
| 256 |
+
"eval_loss": 1.7677279710769653,
|
| 257 |
+
"eval_runtime": 577.7402,
|
| 258 |
+
"eval_samples_per_second": 3.083,
|
| 259 |
+
"eval_steps_per_second": 0.386,
|
| 260 |
+
"step": 82908
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 19.0,
|
| 264 |
+
"learning_rate": 2.403473729917499e-06,
|
| 265 |
+
"loss": 1.6696,
|
| 266 |
+
"step": 87514
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 19.0,
|
| 270 |
+
"eval_loss": 1.770689606666565,
|
| 271 |
+
"eval_runtime": 240.6916,
|
| 272 |
+
"eval_samples_per_second": 7.4,
|
| 273 |
+
"eval_steps_per_second": 0.926,
|
| 274 |
+
"step": 87514
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 20.0,
|
| 278 |
+
"learning_rate": 2.003647416413374e-06,
|
| 279 |
+
"loss": 1.6622,
|
| 280 |
+
"step": 92120
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 20.0,
|
| 284 |
+
"eval_loss": 1.7661957740783691,
|
| 285 |
+
"eval_runtime": 241.9531,
|
| 286 |
+
"eval_samples_per_second": 7.361,
|
| 287 |
+
"eval_steps_per_second": 0.922,
|
| 288 |
+
"step": 92120
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 21.0,
|
| 292 |
+
"learning_rate": 1.603821102909249e-06,
|
| 293 |
+
"loss": 1.6554,
|
| 294 |
+
"step": 96726
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 21.0,
|
| 298 |
+
"eval_loss": 1.7594928741455078,
|
| 299 |
+
"eval_runtime": 241.3426,
|
| 300 |
+
"eval_samples_per_second": 7.38,
|
| 301 |
+
"eval_steps_per_second": 0.924,
|
| 302 |
+
"step": 96726
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 22.0,
|
| 306 |
+
"learning_rate": 1.2039079461571863e-06,
|
| 307 |
+
"loss": 1.6529,
|
| 308 |
+
"step": 101332
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 22.0,
|
| 312 |
+
"eval_loss": 1.7429261207580566,
|
| 313 |
+
"eval_runtime": 240.7407,
|
| 314 |
+
"eval_samples_per_second": 7.398,
|
| 315 |
+
"eval_steps_per_second": 0.926,
|
| 316 |
+
"step": 101332
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 23.0,
|
| 320 |
+
"learning_rate": 8.039947894051238e-07,
|
| 321 |
+
"loss": 1.6441,
|
| 322 |
+
"step": 105938
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 23.0,
|
| 326 |
+
"eval_loss": 1.739193081855774,
|
| 327 |
+
"eval_runtime": 239.2341,
|
| 328 |
+
"eval_samples_per_second": 7.445,
|
| 329 |
+
"eval_steps_per_second": 0.932,
|
| 330 |
+
"step": 105938
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 24.0,
|
| 334 |
+
"learning_rate": 4.041684759009987e-07,
|
| 335 |
+
"loss": 1.6395,
|
| 336 |
+
"step": 110544
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 24.0,
|
| 340 |
+
"eval_loss": 1.7394294738769531,
|
| 341 |
+
"eval_runtime": 238.4537,
|
| 342 |
+
"eval_samples_per_second": 7.469,
|
| 343 |
+
"eval_steps_per_second": 0.935,
|
| 344 |
+
"step": 110544
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 25.0,
|
| 348 |
+
"learning_rate": 4.25531914893617e-09,
|
| 349 |
+
"loss": 1.6329,
|
| 350 |
+
"step": 115150
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 25.0,
|
| 354 |
+
"eval_loss": 1.7369647026062012,
|
| 355 |
+
"eval_runtime": 240.7806,
|
| 356 |
+
"eval_samples_per_second": 7.397,
|
| 357 |
+
"eval_steps_per_second": 0.926,
|
| 358 |
+
"step": 115150
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 25.0,
|
| 362 |
+
"step": 115150,
|
| 363 |
+
"total_flos": 2.2008141195206407e+20,
|
| 364 |
+
"train_loss": 1.7444752132069583,
|
| 365 |
+
"train_runtime": 191692.4998,
|
| 366 |
+
"train_samples_per_second": 2.402,
|
| 367 |
+
"train_steps_per_second": 0.601
|
| 368 |
+
}
|
| 369 |
+
],
|
| 370 |
+
"max_steps": 115150,
|
| 371 |
+
"num_train_epochs": 25,
|
| 372 |
+
"total_flos": 2.2008141195206407e+20,
|
| 373 |
+
"trial_name": null,
|
| 374 |
+
"trial_params": null
|
| 375 |
+
}
|