Upload checkpoint-1200/trainer_state.json
Browse files
checkpoint-1200/trainer_state.json
ADDED
|
@@ -0,0 +1,298 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1050,
|
| 3 |
+
"best_metric": 0.014976329170167446,
|
| 4 |
+
"best_model_checkpoint": "/content/drive/MyDrive/model/Roberta_Squad/roberta_qa_squad/checkpoint-1050",
|
| 5 |
+
"epoch": 1.124706985466479,
|
| 6 |
+
"eval_steps": 150,
|
| 7 |
+
"global_step": 1200,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0009376465072667605,
|
| 14 |
+
"grad_norm": 15.076981544494629,
|
| 15 |
+
"learning_rate": 0.0,
|
| 16 |
+
"loss": 1.4739,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.04688232536333802,
|
| 21 |
+
"grad_norm": 14.212701797485352,
|
| 22 |
+
"learning_rate": 3.598875351452672e-07,
|
| 23 |
+
"loss": 1.4366,
|
| 24 |
+
"step": 50
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.09376465072667604,
|
| 28 |
+
"grad_norm": 12.904585838317871,
|
| 29 |
+
"learning_rate": 7.347703842549205e-07,
|
| 30 |
+
"loss": 1.1476,
|
| 31 |
+
"step": 100
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.14064697609001406,
|
| 35 |
+
"grad_norm": 5.23610258102417,
|
| 36 |
+
"learning_rate": 1.1096532333645738e-06,
|
| 37 |
+
"loss": 0.6204,
|
| 38 |
+
"step": 150
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.14064697609001406,
|
| 42 |
+
"eval_accuracy": 0.9943728018757327,
|
| 43 |
+
"eval_f1": 0.9915671414895327,
|
| 44 |
+
"eval_loss": 0.2603488862514496,
|
| 45 |
+
"eval_runtime": 77.8283,
|
| 46 |
+
"eval_samples_per_second": 109.6,
|
| 47 |
+
"eval_steps_per_second": 3.431,
|
| 48 |
+
"step": 150
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.1875293014533521,
|
| 52 |
+
"grad_norm": 1.283129096031189,
|
| 53 |
+
"learning_rate": 1.484536082474227e-06,
|
| 54 |
+
"loss": 0.1789,
|
| 55 |
+
"step": 200
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.23441162681669012,
|
| 59 |
+
"grad_norm": 0.6366870403289795,
|
| 60 |
+
"learning_rate": 1.8594189315838805e-06,
|
| 61 |
+
"loss": 0.066,
|
| 62 |
+
"step": 250
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.2812939521800281,
|
| 66 |
+
"grad_norm": 0.34226417541503906,
|
| 67 |
+
"learning_rate": 2.2343017806935337e-06,
|
| 68 |
+
"loss": 0.0497,
|
| 69 |
+
"step": 300
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.2812939521800281,
|
| 73 |
+
"eval_accuracy": 0.9943728018757327,
|
| 74 |
+
"eval_f1": 0.9915671414895327,
|
| 75 |
+
"eval_loss": 0.04236311838030815,
|
| 76 |
+
"eval_runtime": 77.3474,
|
| 77 |
+
"eval_samples_per_second": 110.282,
|
| 78 |
+
"eval_steps_per_second": 3.452,
|
| 79 |
+
"step": 300
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.32817627754336615,
|
| 83 |
+
"grad_norm": 1.778275489807129,
|
| 84 |
+
"learning_rate": 2.6091846298031867e-06,
|
| 85 |
+
"loss": 0.059,
|
| 86 |
+
"step": 350
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.3750586029067042,
|
| 90 |
+
"grad_norm": 0.3131299316883087,
|
| 91 |
+
"learning_rate": 2.98406747891284e-06,
|
| 92 |
+
"loss": 0.0414,
|
| 93 |
+
"step": 400
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.4219409282700422,
|
| 97 |
+
"grad_norm": 0.203192800283432,
|
| 98 |
+
"learning_rate": 3.3589503280224936e-06,
|
| 99 |
+
"loss": 0.0385,
|
| 100 |
+
"step": 450
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.4219409282700422,
|
| 104 |
+
"eval_accuracy": 0.9943728018757327,
|
| 105 |
+
"eval_f1": 0.9915671414895327,
|
| 106 |
+
"eval_loss": 0.027930891141295433,
|
| 107 |
+
"eval_runtime": 77.8018,
|
| 108 |
+
"eval_samples_per_second": 109.638,
|
| 109 |
+
"eval_steps_per_second": 3.432,
|
| 110 |
+
"step": 450
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.46882325363338023,
|
| 114 |
+
"grad_norm": 0.5342408418655396,
|
| 115 |
+
"learning_rate": 3.7338331771321466e-06,
|
| 116 |
+
"loss": 0.0392,
|
| 117 |
+
"step": 500
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.5157055789967182,
|
| 121 |
+
"grad_norm": 0.7715256214141846,
|
| 122 |
+
"learning_rate": 4.1087160262418e-06,
|
| 123 |
+
"loss": 0.0303,
|
| 124 |
+
"step": 550
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.5625879043600562,
|
| 128 |
+
"grad_norm": 2.367220163345337,
|
| 129 |
+
"learning_rate": 4.4835988753514534e-06,
|
| 130 |
+
"loss": 0.0208,
|
| 131 |
+
"step": 600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.5625879043600562,
|
| 135 |
+
"eval_accuracy": 0.9967174677608441,
|
| 136 |
+
"eval_f1": 0.9958795497944622,
|
| 137 |
+
"eval_loss": 0.018059343099594116,
|
| 138 |
+
"eval_runtime": 77.6862,
|
| 139 |
+
"eval_samples_per_second": 109.801,
|
| 140 |
+
"eval_steps_per_second": 3.437,
|
| 141 |
+
"step": 600
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.6094702297233943,
|
| 145 |
+
"grad_norm": 5.750198841094971,
|
| 146 |
+
"learning_rate": 4.8584817244611065e-06,
|
| 147 |
+
"loss": 0.0211,
|
| 148 |
+
"step": 650
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.6563525550867323,
|
| 152 |
+
"grad_norm": 1.2362596988677979,
|
| 153 |
+
"learning_rate": 5.2333645735707595e-06,
|
| 154 |
+
"loss": 0.0262,
|
| 155 |
+
"step": 700
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.7032348804500703,
|
| 159 |
+
"grad_norm": 0.045022640377283096,
|
| 160 |
+
"learning_rate": 5.608247422680413e-06,
|
| 161 |
+
"loss": 0.0142,
|
| 162 |
+
"step": 750
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.7032348804500703,
|
| 166 |
+
"eval_accuracy": 0.9963657678780774,
|
| 167 |
+
"eval_f1": 0.9953991663366644,
|
| 168 |
+
"eval_loss": 0.0179133377969265,
|
| 169 |
+
"eval_runtime": 78.1703,
|
| 170 |
+
"eval_samples_per_second": 109.121,
|
| 171 |
+
"eval_steps_per_second": 3.416,
|
| 172 |
+
"step": 750
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.7501172058134083,
|
| 176 |
+
"grad_norm": 0.2589455544948578,
|
| 177 |
+
"learning_rate": 5.983130271790066e-06,
|
| 178 |
+
"loss": 0.0229,
|
| 179 |
+
"step": 800
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.7969995311767464,
|
| 183 |
+
"grad_norm": 1.0219216346740723,
|
| 184 |
+
"learning_rate": 6.35801312089972e-06,
|
| 185 |
+
"loss": 0.0272,
|
| 186 |
+
"step": 850
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.8438818565400844,
|
| 190 |
+
"grad_norm": 5.783281326293945,
|
| 191 |
+
"learning_rate": 6.732895970009373e-06,
|
| 192 |
+
"loss": 0.029,
|
| 193 |
+
"step": 900
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.8438818565400844,
|
| 197 |
+
"eval_accuracy": 0.9964830011723329,
|
| 198 |
+
"eval_f1": 0.9955380069992853,
|
| 199 |
+
"eval_loss": 0.015631183981895447,
|
| 200 |
+
"eval_runtime": 77.5242,
|
| 201 |
+
"eval_samples_per_second": 110.03,
|
| 202 |
+
"eval_steps_per_second": 3.444,
|
| 203 |
+
"step": 900
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 0.8907641819034224,
|
| 207 |
+
"grad_norm": 0.03242630138993263,
|
| 208 |
+
"learning_rate": 7.107778819119027e-06,
|
| 209 |
+
"loss": 0.0136,
|
| 210 |
+
"step": 950
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 0.9376465072667605,
|
| 214 |
+
"grad_norm": 0.07519914209842682,
|
| 215 |
+
"learning_rate": 7.48266166822868e-06,
|
| 216 |
+
"loss": 0.0167,
|
| 217 |
+
"step": 1000
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.9845288326300985,
|
| 221 |
+
"grad_norm": 8.501079559326172,
|
| 222 |
+
"learning_rate": 7.857544517338333e-06,
|
| 223 |
+
"loss": 0.0153,
|
| 224 |
+
"step": 1050
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.9845288326300985,
|
| 228 |
+
"eval_accuracy": 0.9963657678780774,
|
| 229 |
+
"eval_f1": 0.9954965041531265,
|
| 230 |
+
"eval_loss": 0.014976329170167446,
|
| 231 |
+
"eval_runtime": 77.6596,
|
| 232 |
+
"eval_samples_per_second": 109.838,
|
| 233 |
+
"eval_steps_per_second": 3.438,
|
| 234 |
+
"step": 1050
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.030942334739803,
|
| 238 |
+
"grad_norm": 0.6377618908882141,
|
| 239 |
+
"learning_rate": 8.232427366447986e-06,
|
| 240 |
+
"loss": 0.0152,
|
| 241 |
+
"step": 1100
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.0778246601031412,
|
| 245 |
+
"grad_norm": 0.05754481628537178,
|
| 246 |
+
"learning_rate": 8.607310215557639e-06,
|
| 247 |
+
"loss": 0.0225,
|
| 248 |
+
"step": 1150
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.124706985466479,
|
| 252 |
+
"grad_norm": 0.14188943803310394,
|
| 253 |
+
"learning_rate": 8.982193064667292e-06,
|
| 254 |
+
"loss": 0.0163,
|
| 255 |
+
"step": 1200
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.124706985466479,
|
| 259 |
+
"eval_accuracy": 0.9963657678780774,
|
| 260 |
+
"eval_f1": 0.9954437982952228,
|
| 261 |
+
"eval_loss": 0.017581254243850708,
|
| 262 |
+
"eval_runtime": 77.8986,
|
| 263 |
+
"eval_samples_per_second": 109.501,
|
| 264 |
+
"eval_steps_per_second": 3.428,
|
| 265 |
+
"step": 1200
|
| 266 |
+
}
|
| 267 |
+
],
|
| 268 |
+
"logging_steps": 50,
|
| 269 |
+
"max_steps": 10670,
|
| 270 |
+
"num_input_tokens_seen": 0,
|
| 271 |
+
"num_train_epochs": 10,
|
| 272 |
+
"save_steps": 150,
|
| 273 |
+
"stateful_callbacks": {
|
| 274 |
+
"EarlyStoppingCallback": {
|
| 275 |
+
"args": {
|
| 276 |
+
"early_stopping_patience": 4,
|
| 277 |
+
"early_stopping_threshold": 0.001
|
| 278 |
+
},
|
| 279 |
+
"attributes": {
|
| 280 |
+
"early_stopping_patience_counter": 2
|
| 281 |
+
}
|
| 282 |
+
},
|
| 283 |
+
"TrainerControl": {
|
| 284 |
+
"args": {
|
| 285 |
+
"should_epoch_stop": false,
|
| 286 |
+
"should_evaluate": false,
|
| 287 |
+
"should_log": false,
|
| 288 |
+
"should_save": true,
|
| 289 |
+
"should_training_stop": false
|
| 290 |
+
},
|
| 291 |
+
"attributes": {}
|
| 292 |
+
}
|
| 293 |
+
},
|
| 294 |
+
"total_flos": 2.019484367039693e+16,
|
| 295 |
+
"train_batch_size": 32,
|
| 296 |
+
"trial_name": null,
|
| 297 |
+
"trial_params": null
|
| 298 |
+
}
|