End of training
Browse files- all_results.json +15 -0
- eval_results.json +10 -0
- train_results.json +8 -0
- trainer_state.json +289 -0
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_cer": 0.5038576602109904,
|
| 4 |
+
"eval_loss": 2.1666207313537598,
|
| 5 |
+
"eval_runtime": 25.4631,
|
| 6 |
+
"eval_samples": 456,
|
| 7 |
+
"eval_samples_per_second": 17.908,
|
| 8 |
+
"eval_steps_per_second": 2.239,
|
| 9 |
+
"eval_wer": 0.9737239504681365,
|
| 10 |
+
"train_loss": 8.309904957913805,
|
| 11 |
+
"train_runtime": 17807.6517,
|
| 12 |
+
"train_samples": 22262,
|
| 13 |
+
"train_samples_per_second": 12.501,
|
| 14 |
+
"train_steps_per_second": 0.195
|
| 15 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_cer": 0.5038576602109904,
|
| 4 |
+
"eval_loss": 2.1666207313537598,
|
| 5 |
+
"eval_runtime": 25.4631,
|
| 6 |
+
"eval_samples": 456,
|
| 7 |
+
"eval_samples_per_second": 17.908,
|
| 8 |
+
"eval_steps_per_second": 2.239,
|
| 9 |
+
"eval_wer": 0.9737239504681365
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 8.309904957913805,
|
| 4 |
+
"train_runtime": 17807.6517,
|
| 5 |
+
"train_samples": 22262,
|
| 6 |
+
"train_samples_per_second": 12.501,
|
| 7 |
+
"train_steps_per_second": 0.195
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,289 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 3480,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.29,
|
| 12 |
+
"learning_rate": 3.7499999999999997e-06,
|
| 13 |
+
"loss": 47.1781,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.57,
|
| 18 |
+
"learning_rate": 7.499999999999999e-06,
|
| 19 |
+
"loss": 33.2186,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.86,
|
| 24 |
+
"learning_rate": 1.1249999999999999e-05,
|
| 25 |
+
"loss": 26.4407,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.15,
|
| 30 |
+
"learning_rate": 1.4999999999999999e-05,
|
| 31 |
+
"loss": 23.1487,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.44,
|
| 36 |
+
"learning_rate": 1.875e-05,
|
| 37 |
+
"loss": 19.558,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.44,
|
| 42 |
+
"eval_cer": 1.0,
|
| 43 |
+
"eval_loss": 19.4094295501709,
|
| 44 |
+
"eval_runtime": 25.117,
|
| 45 |
+
"eval_samples_per_second": 18.155,
|
| 46 |
+
"eval_steps_per_second": 2.269,
|
| 47 |
+
"eval_wer": 1.0,
|
| 48 |
+
"step": 500
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 1.72,
|
| 52 |
+
"learning_rate": 2.2499999999999998e-05,
|
| 53 |
+
"loss": 15.5015,
|
| 54 |
+
"step": 600
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 2.01,
|
| 58 |
+
"learning_rate": 2.6249999999999998e-05,
|
| 59 |
+
"loss": 11.0351,
|
| 60 |
+
"step": 700
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.3,
|
| 64 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 65 |
+
"loss": 7.2384,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 2.59,
|
| 70 |
+
"learning_rate": 3.375e-05,
|
| 71 |
+
"loss": 5.2354,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 2.87,
|
| 76 |
+
"learning_rate": 3.75e-05,
|
| 77 |
+
"loss": 4.7968,
|
| 78 |
+
"step": 1000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 2.87,
|
| 82 |
+
"eval_cer": 1.0,
|
| 83 |
+
"eval_loss": 4.782816410064697,
|
| 84 |
+
"eval_runtime": 25.1318,
|
| 85 |
+
"eval_samples_per_second": 18.144,
|
| 86 |
+
"eval_steps_per_second": 2.268,
|
| 87 |
+
"eval_wer": 1.0,
|
| 88 |
+
"step": 1000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 3.16,
|
| 92 |
+
"learning_rate": 4.125e-05,
|
| 93 |
+
"loss": 4.7437,
|
| 94 |
+
"step": 1100
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 3.45,
|
| 98 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 99 |
+
"loss": 4.654,
|
| 100 |
+
"step": 1200
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 3.74,
|
| 104 |
+
"learning_rate": 4.875e-05,
|
| 105 |
+
"loss": 4.6017,
|
| 106 |
+
"step": 1300
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 4.02,
|
| 110 |
+
"learning_rate": 5.2499999999999995e-05,
|
| 111 |
+
"loss": 4.5759,
|
| 112 |
+
"step": 1400
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 4.31,
|
| 116 |
+
"learning_rate": 5.625e-05,
|
| 117 |
+
"loss": 4.5125,
|
| 118 |
+
"step": 1500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 4.31,
|
| 122 |
+
"eval_cer": 0.9540229885057471,
|
| 123 |
+
"eval_loss": 4.495882987976074,
|
| 124 |
+
"eval_runtime": 25.1686,
|
| 125 |
+
"eval_samples_per_second": 18.118,
|
| 126 |
+
"eval_steps_per_second": 2.265,
|
| 127 |
+
"eval_wer": 0.9990939293264874,
|
| 128 |
+
"step": 1500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 4.6,
|
| 132 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 133 |
+
"loss": 4.3867,
|
| 134 |
+
"step": 1600
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 4.89,
|
| 138 |
+
"learning_rate": 6.374999999999999e-05,
|
| 139 |
+
"loss": 4.304,
|
| 140 |
+
"step": 1700
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 5.17,
|
| 144 |
+
"learning_rate": 6.75e-05,
|
| 145 |
+
"loss": 4.261,
|
| 146 |
+
"step": 1800
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 5.46,
|
| 150 |
+
"learning_rate": 7.125e-05,
|
| 151 |
+
"loss": 4.2279,
|
| 152 |
+
"step": 1900
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 5.75,
|
| 156 |
+
"learning_rate": 7.5e-05,
|
| 157 |
+
"loss": 4.2202,
|
| 158 |
+
"step": 2000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 5.75,
|
| 162 |
+
"eval_cer": 0.8520311761927255,
|
| 163 |
+
"eval_loss": 4.290519714355469,
|
| 164 |
+
"eval_runtime": 24.9956,
|
| 165 |
+
"eval_samples_per_second": 18.243,
|
| 166 |
+
"eval_steps_per_second": 2.28,
|
| 167 |
+
"eval_wer": 0.9922983992751434,
|
| 168 |
+
"step": 2000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.03,
|
| 172 |
+
"learning_rate": 6.993243243243243e-05,
|
| 173 |
+
"loss": 4.184,
|
| 174 |
+
"step": 2100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 6.32,
|
| 178 |
+
"learning_rate": 6.486486486486486e-05,
|
| 179 |
+
"loss": 4.1509,
|
| 180 |
+
"step": 2200
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 6.61,
|
| 184 |
+
"learning_rate": 5.979729729729729e-05,
|
| 185 |
+
"loss": 4.0915,
|
| 186 |
+
"step": 2300
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 6.9,
|
| 190 |
+
"learning_rate": 5.4729729729729724e-05,
|
| 191 |
+
"loss": 3.9577,
|
| 192 |
+
"step": 2400
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 7.18,
|
| 196 |
+
"learning_rate": 4.966216216216216e-05,
|
| 197 |
+
"loss": 3.7774,
|
| 198 |
+
"step": 2500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 7.18,
|
| 202 |
+
"eval_cer": 0.6651708392379153,
|
| 203 |
+
"eval_loss": 3.2845523357391357,
|
| 204 |
+
"eval_runtime": 25.0256,
|
| 205 |
+
"eval_samples_per_second": 18.221,
|
| 206 |
+
"eval_steps_per_second": 2.278,
|
| 207 |
+
"eval_wer": 1.0356387798248263,
|
| 208 |
+
"step": 2500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 7.47,
|
| 212 |
+
"learning_rate": 4.4594594594594596e-05,
|
| 213 |
+
"loss": 3.601,
|
| 214 |
+
"step": 2600
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 7.76,
|
| 218 |
+
"learning_rate": 3.952702702702702e-05,
|
| 219 |
+
"loss": 3.4439,
|
| 220 |
+
"step": 2700
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 8.05,
|
| 224 |
+
"learning_rate": 3.4459459459459455e-05,
|
| 225 |
+
"loss": 3.3193,
|
| 226 |
+
"step": 2800
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.33,
|
| 230 |
+
"learning_rate": 2.9391891891891888e-05,
|
| 231 |
+
"loss": 3.2091,
|
| 232 |
+
"step": 2900
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 8.62,
|
| 236 |
+
"learning_rate": 2.4324324324324324e-05,
|
| 237 |
+
"loss": 3.1418,
|
| 238 |
+
"step": 3000
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 8.62,
|
| 242 |
+
"eval_cer": 0.5429459927570461,
|
| 243 |
+
"eval_loss": 2.362445592880249,
|
| 244 |
+
"eval_runtime": 25.1664,
|
| 245 |
+
"eval_samples_per_second": 18.119,
|
| 246 |
+
"eval_steps_per_second": 2.265,
|
| 247 |
+
"eval_wer": 0.988221081244337,
|
| 248 |
+
"step": 3000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.91,
|
| 252 |
+
"learning_rate": 1.9256756756756753e-05,
|
| 253 |
+
"loss": 3.0785,
|
| 254 |
+
"step": 3100
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 9.2,
|
| 258 |
+
"learning_rate": 1.4189189189189189e-05,
|
| 259 |
+
"loss": 3.0369,
|
| 260 |
+
"step": 3200
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 9.48,
|
| 264 |
+
"learning_rate": 9.121621621621622e-06,
|
| 265 |
+
"loss": 3.003,
|
| 266 |
+
"step": 3300
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 9.77,
|
| 270 |
+
"learning_rate": 4.054054054054054e-06,
|
| 271 |
+
"loss": 2.977,
|
| 272 |
+
"step": 3400
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 10.0,
|
| 276 |
+
"step": 3480,
|
| 277 |
+
"total_flos": 5.681228030880589e+19,
|
| 278 |
+
"train_loss": 8.309904957913805,
|
| 279 |
+
"train_runtime": 17807.6517,
|
| 280 |
+
"train_samples_per_second": 12.501,
|
| 281 |
+
"train_steps_per_second": 0.195
|
| 282 |
+
}
|
| 283 |
+
],
|
| 284 |
+
"max_steps": 3480,
|
| 285 |
+
"num_train_epochs": 10,
|
| 286 |
+
"total_flos": 5.681228030880589e+19,
|
| 287 |
+
"trial_name": null,
|
| 288 |
+
"trial_params": null
|
| 289 |
+
}
|