Commit
•
bab95ca
1
Parent(s):
355a1bd
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.11277494579553604,
|
| 4 |
+
"eval_runtime": 182.7513,
|
| 5 |
+
"eval_samples": 3999,
|
| 6 |
+
"eval_samples_per_second": 21.882,
|
| 7 |
+
"eval_steps_per_second": 2.736,
|
| 8 |
+
"eval_wer": 0.09472401608383088,
|
| 9 |
+
"train_loss": 0.4369275619506836,
|
| 10 |
+
"train_runtime": 9373.6438,
|
| 11 |
+
"train_samples": 5999,
|
| 12 |
+
"train_samples_per_second": 6.4,
|
| 13 |
+
"train_steps_per_second": 0.8
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.11277494579553604,
|
| 4 |
+
"eval_runtime": 182.7513,
|
| 5 |
+
"eval_samples": 3999,
|
| 6 |
+
"eval_samples_per_second": 21.882,
|
| 7 |
+
"eval_steps_per_second": 2.736,
|
| 8 |
+
"eval_wer": 0.09472401608383088
|
| 9 |
+
}
|
runs/May23_12-38-18_DESKTOP-ASUS/events.out.tfevents.1684848656.DESKTOP-ASUS.17550.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:147891e882c2be2f7cb41de785bf3155e7cf1e09833be498bd22915088975057
|
| 3 |
+
size 358
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 0.4369275619506836,
|
| 4 |
+
"train_runtime": 9373.6438,
|
| 5 |
+
"train_samples": 5999,
|
| 6 |
+
"train_samples_per_second": 6.4,
|
| 7 |
+
"train_steps_per_second": 0.8
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,277 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 7500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.53,
|
| 12 |
+
"eval_loss": 1.229895830154419,
|
| 13 |
+
"eval_runtime": 183.9271,
|
| 14 |
+
"eval_samples_per_second": 21.742,
|
| 15 |
+
"eval_steps_per_second": 2.718,
|
| 16 |
+
"eval_wer": 0.7600584866577312,
|
| 17 |
+
"step": 400
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.67,
|
| 21 |
+
"learning_rate": 0.0002982,
|
| 22 |
+
"loss": 3.6201,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 1.07,
|
| 27 |
+
"eval_loss": 0.49728861451148987,
|
| 28 |
+
"eval_runtime": 182.3086,
|
| 29 |
+
"eval_samples_per_second": 21.935,
|
| 30 |
+
"eval_steps_per_second": 2.743,
|
| 31 |
+
"eval_wer": 0.42488119897648347,
|
| 32 |
+
"step": 800
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.33,
|
| 36 |
+
"learning_rate": 0.0002787,
|
| 37 |
+
"loss": 0.6369,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.6,
|
| 42 |
+
"eval_loss": 0.3747561573982239,
|
| 43 |
+
"eval_runtime": 183.2249,
|
| 44 |
+
"eval_samples_per_second": 21.826,
|
| 45 |
+
"eval_steps_per_second": 2.729,
|
| 46 |
+
"eval_wer": 0.3480930912635555,
|
| 47 |
+
"step": 1200
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 2.0,
|
| 51 |
+
"learning_rate": 0.00025727142857142856,
|
| 52 |
+
"loss": 0.455,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.13,
|
| 57 |
+
"eval_loss": 0.28336068987846375,
|
| 58 |
+
"eval_runtime": 182.904,
|
| 59 |
+
"eval_samples_per_second": 21.864,
|
| 60 |
+
"eval_steps_per_second": 2.734,
|
| 61 |
+
"eval_wer": 0.2643840623857682,
|
| 62 |
+
"step": 1600
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.67,
|
| 66 |
+
"learning_rate": 0.00023584285714285712,
|
| 67 |
+
"loss": 0.3177,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.67,
|
| 72 |
+
"eval_loss": 0.24264603853225708,
|
| 73 |
+
"eval_runtime": 183.432,
|
| 74 |
+
"eval_samples_per_second": 21.801,
|
| 75 |
+
"eval_steps_per_second": 2.726,
|
| 76 |
+
"eval_wer": 0.22339466309248202,
|
| 77 |
+
"step": 2000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 3.2,
|
| 81 |
+
"eval_loss": 0.18675319850444794,
|
| 82 |
+
"eval_runtime": 183.8214,
|
| 83 |
+
"eval_samples_per_second": 21.755,
|
| 84 |
+
"eval_steps_per_second": 2.72,
|
| 85 |
+
"eval_wer": 0.18615815767028146,
|
| 86 |
+
"step": 2400
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.33,
|
| 90 |
+
"learning_rate": 0.0002144142857142857,
|
| 91 |
+
"loss": 0.2697,
|
| 92 |
+
"step": 2500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.73,
|
| 96 |
+
"eval_loss": 0.19151142239570618,
|
| 97 |
+
"eval_runtime": 183.9907,
|
| 98 |
+
"eval_samples_per_second": 21.735,
|
| 99 |
+
"eval_steps_per_second": 2.718,
|
| 100 |
+
"eval_wer": 0.18472036066772268,
|
| 101 |
+
"step": 2800
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 4.0,
|
| 105 |
+
"learning_rate": 0.00019298571428571424,
|
| 106 |
+
"loss": 0.2363,
|
| 107 |
+
"step": 3000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 4.27,
|
| 111 |
+
"eval_loss": 0.166675865650177,
|
| 112 |
+
"eval_runtime": 184.6,
|
| 113 |
+
"eval_samples_per_second": 21.663,
|
| 114 |
+
"eval_steps_per_second": 2.709,
|
| 115 |
+
"eval_wer": 0.16083830876081395,
|
| 116 |
+
"step": 3200
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 4.67,
|
| 120 |
+
"learning_rate": 0.00017155714285714283,
|
| 121 |
+
"loss": 0.1795,
|
| 122 |
+
"step": 3500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 4.8,
|
| 126 |
+
"eval_loss": 0.1457526981830597,
|
| 127 |
+
"eval_runtime": 183.1764,
|
| 128 |
+
"eval_samples_per_second": 21.831,
|
| 129 |
+
"eval_steps_per_second": 2.73,
|
| 130 |
+
"eval_wer": 0.14287803094918972,
|
| 131 |
+
"step": 3600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.33,
|
| 135 |
+
"learning_rate": 0.00015012857142857142,
|
| 136 |
+
"loss": 0.1636,
|
| 137 |
+
"step": 4000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 5.33,
|
| 141 |
+
"eval_loss": 0.14684966206550598,
|
| 142 |
+
"eval_runtime": 182.9497,
|
| 143 |
+
"eval_samples_per_second": 21.858,
|
| 144 |
+
"eval_steps_per_second": 2.733,
|
| 145 |
+
"eval_wer": 0.1388083343487267,
|
| 146 |
+
"step": 4000
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 5.87,
|
| 150 |
+
"eval_loss": 0.13510963320732117,
|
| 151 |
+
"eval_runtime": 182.9435,
|
| 152 |
+
"eval_samples_per_second": 21.859,
|
| 153 |
+
"eval_steps_per_second": 2.733,
|
| 154 |
+
"eval_wer": 0.1314487632508834,
|
| 155 |
+
"step": 4400
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 6.0,
|
| 159 |
+
"learning_rate": 0.00012869999999999998,
|
| 160 |
+
"loss": 0.1445,
|
| 161 |
+
"step": 4500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 6.4,
|
| 165 |
+
"eval_loss": 0.11629413068294525,
|
| 166 |
+
"eval_runtime": 183.428,
|
| 167 |
+
"eval_samples_per_second": 21.801,
|
| 168 |
+
"eval_steps_per_second": 2.726,
|
| 169 |
+
"eval_wer": 0.11078347751919093,
|
| 170 |
+
"step": 4800
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 6.67,
|
| 174 |
+
"learning_rate": 0.00010727142857142857,
|
| 175 |
+
"loss": 0.1153,
|
| 176 |
+
"step": 5000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.93,
|
| 180 |
+
"eval_loss": 0.10931886732578278,
|
| 181 |
+
"eval_runtime": 183.2606,
|
| 182 |
+
"eval_samples_per_second": 21.821,
|
| 183 |
+
"eval_steps_per_second": 2.728,
|
| 184 |
+
"eval_wer": 0.10876081393932009,
|
| 185 |
+
"step": 5200
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 7.33,
|
| 189 |
+
"learning_rate": 8.584285714285713e-05,
|
| 190 |
+
"loss": 0.1011,
|
| 191 |
+
"step": 5500
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.47,
|
| 195 |
+
"eval_loss": 0.12333084642887115,
|
| 196 |
+
"eval_runtime": 183.8962,
|
| 197 |
+
"eval_samples_per_second": 21.746,
|
| 198 |
+
"eval_steps_per_second": 2.719,
|
| 199 |
+
"eval_wer": 0.11409772145729255,
|
| 200 |
+
"step": 5600
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.0,
|
| 204 |
+
"learning_rate": 6.441428571428571e-05,
|
| 205 |
+
"loss": 0.0978,
|
| 206 |
+
"step": 6000
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 8.0,
|
| 210 |
+
"eval_loss": 0.1147131398320198,
|
| 211 |
+
"eval_runtime": 183.2548,
|
| 212 |
+
"eval_samples_per_second": 21.822,
|
| 213 |
+
"eval_steps_per_second": 2.728,
|
| 214 |
+
"eval_wer": 0.10408188132082369,
|
| 215 |
+
"step": 6000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 8.53,
|
| 219 |
+
"eval_loss": 0.11119665205478668,
|
| 220 |
+
"eval_runtime": 183.6945,
|
| 221 |
+
"eval_samples_per_second": 21.77,
|
| 222 |
+
"eval_steps_per_second": 2.722,
|
| 223 |
+
"eval_wer": 0.09840380163275253,
|
| 224 |
+
"step": 6400
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 8.67,
|
| 228 |
+
"learning_rate": 4.2985714285714285e-05,
|
| 229 |
+
"loss": 0.0748,
|
| 230 |
+
"step": 6500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 9.07,
|
| 234 |
+
"eval_loss": 0.11000872403383255,
|
| 235 |
+
"eval_runtime": 184.0303,
|
| 236 |
+
"eval_samples_per_second": 21.73,
|
| 237 |
+
"eval_steps_per_second": 2.717,
|
| 238 |
+
"eval_wer": 0.09382234677714146,
|
| 239 |
+
"step": 6800
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 9.33,
|
| 243 |
+
"learning_rate": 2.1557142857142854e-05,
|
| 244 |
+
"loss": 0.075,
|
| 245 |
+
"step": 7000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 9.6,
|
| 249 |
+
"eval_loss": 0.1180349737405777,
|
| 250 |
+
"eval_runtime": 183.7553,
|
| 251 |
+
"eval_samples_per_second": 21.763,
|
| 252 |
+
"eval_steps_per_second": 2.721,
|
| 253 |
+
"eval_wer": 0.09601559644206166,
|
| 254 |
+
"step": 7200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 10.0,
|
| 258 |
+
"learning_rate": 1.7142857142857143e-07,
|
| 259 |
+
"loss": 0.0668,
|
| 260 |
+
"step": 7500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 10.0,
|
| 264 |
+
"step": 7500,
|
| 265 |
+
"total_flos": 7.948903600818893e+18,
|
| 266 |
+
"train_loss": 0.4369275619506836,
|
| 267 |
+
"train_runtime": 9373.6438,
|
| 268 |
+
"train_samples_per_second": 6.4,
|
| 269 |
+
"train_steps_per_second": 0.8
|
| 270 |
+
}
|
| 271 |
+
],
|
| 272 |
+
"max_steps": 7500,
|
| 273 |
+
"num_train_epochs": 10,
|
| 274 |
+
"total_flos": 7.948903600818893e+18,
|
| 275 |
+
"trial_name": null,
|
| 276 |
+
"trial_params": null
|
| 277 |
+
}
|