Commit
•
9bb8267
1
Parent(s):
4002d6e
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.36,
|
| 3 |
+
"eval_loss": 0.04755961894989014,
|
| 4 |
+
"eval_runtime": 556.6562,
|
| 5 |
+
"eval_samples": 1957,
|
| 6 |
+
"eval_samples_per_second": 3.516,
|
| 7 |
+
"eval_steps_per_second": 0.88,
|
| 8 |
+
"eval_wer": 0.019755170958210215,
|
| 9 |
+
"train_loss": 0.1852580333749453,
|
| 10 |
+
"train_runtime": 12301.5592,
|
| 11 |
+
"train_samples": 8140,
|
| 12 |
+
"train_samples_per_second": 1.561,
|
| 13 |
+
"train_steps_per_second": 0.049
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.36,
|
| 3 |
+
"eval_loss": 0.04755961894989014,
|
| 4 |
+
"eval_runtime": 556.6562,
|
| 5 |
+
"eval_samples": 1957,
|
| 6 |
+
"eval_samples_per_second": 3.516,
|
| 7 |
+
"eval_steps_per_second": 0.88,
|
| 8 |
+
"eval_wer": 0.019755170958210215
|
| 9 |
+
}
|
runs/May15_07-40-47_8388a404bb57/events.out.tfevents.1684149859.8388a404bb57.933.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa4edbbe0f28035a7856d91f1686927179b2d28b63a7271deb65ab25a94891fb
|
| 3 |
+
size 358
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.36,
|
| 3 |
+
"train_loss": 0.1852580333749453,
|
| 4 |
+
"train_runtime": 12301.5592,
|
| 5 |
+
"train_samples": 8140,
|
| 6 |
+
"train_samples_per_second": 1.561,
|
| 7 |
+
"train_steps_per_second": 0.049
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,277 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.3587223587223587,
|
| 5 |
+
"global_step": 600,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.1,
|
| 12 |
+
"learning_rate": 1.5333333333333334e-06,
|
| 13 |
+
"loss": 2.3001,
|
| 14 |
+
"step": 25
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.2,
|
| 18 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 19 |
+
"loss": 0.8218,
|
| 20 |
+
"step": 50
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.2,
|
| 24 |
+
"eval_loss": 0.3785243332386017,
|
| 25 |
+
"eval_runtime": 583.9852,
|
| 26 |
+
"eval_samples_per_second": 3.351,
|
| 27 |
+
"eval_steps_per_second": 0.839,
|
| 28 |
+
"eval_wer": 0.14508231321232587,
|
| 29 |
+
"step": 50
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.29,
|
| 33 |
+
"learning_rate": 4.800000000000001e-06,
|
| 34 |
+
"loss": 0.2521,
|
| 35 |
+
"step": 75
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.39,
|
| 39 |
+
"learning_rate": 6.466666666666667e-06,
|
| 40 |
+
"loss": 0.1429,
|
| 41 |
+
"step": 100
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.39,
|
| 45 |
+
"eval_loss": 0.12126336246728897,
|
| 46 |
+
"eval_runtime": 580.8543,
|
| 47 |
+
"eval_samples_per_second": 3.369,
|
| 48 |
+
"eval_steps_per_second": 0.844,
|
| 49 |
+
"eval_wer": 0.07142254115660616,
|
| 50 |
+
"step": 100
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.49,
|
| 54 |
+
"learning_rate": 8.133333333333334e-06,
|
| 55 |
+
"loss": 0.1167,
|
| 56 |
+
"step": 125
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.59,
|
| 60 |
+
"learning_rate": 9.800000000000001e-06,
|
| 61 |
+
"loss": 0.1155,
|
| 62 |
+
"step": 150
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.59,
|
| 66 |
+
"eval_loss": 0.08073808252811432,
|
| 67 |
+
"eval_runtime": 580.5596,
|
| 68 |
+
"eval_samples_per_second": 3.371,
|
| 69 |
+
"eval_steps_per_second": 0.844,
|
| 70 |
+
"eval_wer": 0.05166737019839595,
|
| 71 |
+
"step": 150
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.69,
|
| 75 |
+
"learning_rate": 9.511111111111112e-06,
|
| 76 |
+
"loss": 0.0949,
|
| 77 |
+
"step": 175
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.79,
|
| 81 |
+
"learning_rate": 8.955555555555555e-06,
|
| 82 |
+
"loss": 0.0764,
|
| 83 |
+
"step": 200
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.79,
|
| 87 |
+
"eval_loss": 0.06518173962831497,
|
| 88 |
+
"eval_runtime": 562.2496,
|
| 89 |
+
"eval_samples_per_second": 3.481,
|
| 90 |
+
"eval_steps_per_second": 0.871,
|
| 91 |
+
"eval_wer": 0.027184466019417475,
|
| 92 |
+
"step": 200
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.88,
|
| 96 |
+
"learning_rate": 8.400000000000001e-06,
|
| 97 |
+
"loss": 0.0707,
|
| 98 |
+
"step": 225
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.98,
|
| 102 |
+
"learning_rate": 7.844444444444446e-06,
|
| 103 |
+
"loss": 0.0724,
|
| 104 |
+
"step": 250
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.98,
|
| 108 |
+
"eval_loss": 0.06074482947587967,
|
| 109 |
+
"eval_runtime": 579.9609,
|
| 110 |
+
"eval_samples_per_second": 3.374,
|
| 111 |
+
"eval_steps_per_second": 0.845,
|
| 112 |
+
"eval_wer": 0.03934149430139299,
|
| 113 |
+
"step": 250
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.08,
|
| 117 |
+
"learning_rate": 7.28888888888889e-06,
|
| 118 |
+
"loss": 0.0401,
|
| 119 |
+
"step": 275
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.18,
|
| 123 |
+
"learning_rate": 6.733333333333334e-06,
|
| 124 |
+
"loss": 0.0357,
|
| 125 |
+
"step": 300
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.18,
|
| 129 |
+
"eval_loss": 0.05690562352538109,
|
| 130 |
+
"eval_runtime": 561.3671,
|
| 131 |
+
"eval_samples_per_second": 3.486,
|
| 132 |
+
"eval_steps_per_second": 0.873,
|
| 133 |
+
"eval_wer": 0.024229632756437317,
|
| 134 |
+
"step": 300
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.28,
|
| 138 |
+
"learning_rate": 6.177777777777778e-06,
|
| 139 |
+
"loss": 0.0345,
|
| 140 |
+
"step": 325
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.38,
|
| 144 |
+
"learning_rate": 5.622222222222222e-06,
|
| 145 |
+
"loss": 0.03,
|
| 146 |
+
"step": 350
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 1.38,
|
| 150 |
+
"eval_loss": 0.055250346660614014,
|
| 151 |
+
"eval_runtime": 564.9558,
|
| 152 |
+
"eval_samples_per_second": 3.464,
|
| 153 |
+
"eval_steps_per_second": 0.867,
|
| 154 |
+
"eval_wer": 0.024314056563951036,
|
| 155 |
+
"step": 350
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 1.47,
|
| 159 |
+
"learning_rate": 5.0666666666666676e-06,
|
| 160 |
+
"loss": 0.0372,
|
| 161 |
+
"step": 375
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 1.57,
|
| 165 |
+
"learning_rate": 4.511111111111111e-06,
|
| 166 |
+
"loss": 0.0325,
|
| 167 |
+
"step": 400
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 1.57,
|
| 171 |
+
"eval_loss": 0.05560004711151123,
|
| 172 |
+
"eval_runtime": 557.3555,
|
| 173 |
+
"eval_samples_per_second": 3.511,
|
| 174 |
+
"eval_steps_per_second": 0.879,
|
| 175 |
+
"eval_wer": 0.022836639932460955,
|
| 176 |
+
"step": 400
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.67,
|
| 180 |
+
"learning_rate": 3.955555555555556e-06,
|
| 181 |
+
"loss": 0.0299,
|
| 182 |
+
"step": 425
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 1.77,
|
| 186 |
+
"learning_rate": 3.4000000000000005e-06,
|
| 187 |
+
"loss": 0.03,
|
| 188 |
+
"step": 450
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.77,
|
| 192 |
+
"eval_loss": 0.050134915858507156,
|
| 193 |
+
"eval_runtime": 559.7287,
|
| 194 |
+
"eval_samples_per_second": 3.496,
|
| 195 |
+
"eval_steps_per_second": 0.875,
|
| 196 |
+
"eval_wer": 0.024187420852680455,
|
| 197 |
+
"step": 450
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 1.87,
|
| 201 |
+
"learning_rate": 2.8444444444444446e-06,
|
| 202 |
+
"loss": 0.0308,
|
| 203 |
+
"step": 475
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.97,
|
| 207 |
+
"learning_rate": 2.2888888888888892e-06,
|
| 208 |
+
"loss": 0.0232,
|
| 209 |
+
"step": 500
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.97,
|
| 213 |
+
"eval_loss": 0.04845602065324783,
|
| 214 |
+
"eval_runtime": 562.6357,
|
| 215 |
+
"eval_samples_per_second": 3.478,
|
| 216 |
+
"eval_steps_per_second": 0.871,
|
| 217 |
+
"eval_wer": 0.020472773322076827,
|
| 218 |
+
"step": 500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.06,
|
| 222 |
+
"learning_rate": 1.7333333333333336e-06,
|
| 223 |
+
"loss": 0.0211,
|
| 224 |
+
"step": 525
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.16,
|
| 228 |
+
"learning_rate": 1.1777777777777778e-06,
|
| 229 |
+
"loss": 0.0143,
|
| 230 |
+
"step": 550
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.16,
|
| 234 |
+
"eval_loss": 0.04800991341471672,
|
| 235 |
+
"eval_runtime": 558.3624,
|
| 236 |
+
"eval_samples_per_second": 3.505,
|
| 237 |
+
"eval_steps_per_second": 0.878,
|
| 238 |
+
"eval_wer": 0.01937526382439848,
|
| 239 |
+
"step": 550
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.26,
|
| 243 |
+
"learning_rate": 6.222222222222223e-07,
|
| 244 |
+
"loss": 0.013,
|
| 245 |
+
"step": 575
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 2.36,
|
| 249 |
+
"learning_rate": 6.666666666666668e-08,
|
| 250 |
+
"loss": 0.0105,
|
| 251 |
+
"step": 600
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.36,
|
| 255 |
+
"eval_loss": 0.04755961894989014,
|
| 256 |
+
"eval_runtime": 557.4658,
|
| 257 |
+
"eval_samples_per_second": 3.511,
|
| 258 |
+
"eval_steps_per_second": 0.879,
|
| 259 |
+
"eval_wer": 0.019755170958210215,
|
| 260 |
+
"step": 600
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.36,
|
| 264 |
+
"step": 600,
|
| 265 |
+
"total_flos": 1.9595614427723858e+19,
|
| 266 |
+
"train_loss": 0.1852580333749453,
|
| 267 |
+
"train_runtime": 12301.5592,
|
| 268 |
+
"train_samples_per_second": 1.561,
|
| 269 |
+
"train_steps_per_second": 0.049
|
| 270 |
+
}
|
| 271 |
+
],
|
| 272 |
+
"max_steps": 600,
|
| 273 |
+
"num_train_epochs": 3,
|
| 274 |
+
"total_flos": 1.9595614427723858e+19,
|
| 275 |
+
"trial_name": null,
|
| 276 |
+
"trial_params": null
|
| 277 |
+
}
|