Commit
•
9b74868
1
Parent(s):
ad304e6
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 99.99,
|
| 3 |
+
"eval_loss": 0.5258437395095825,
|
| 4 |
+
"eval_runtime": 280.7122,
|
| 5 |
+
"eval_samples": 2693,
|
| 6 |
+
"eval_samples_per_second": 9.593,
|
| 7 |
+
"eval_steps_per_second": 0.602,
|
| 8 |
+
"eval_wer": 1.0073082607981607,
|
| 9 |
+
"train_loss": 3.0460862658100742,
|
| 10 |
+
"train_runtime": 81000.8052,
|
| 11 |
+
"train_samples": 4001,
|
| 12 |
+
"train_samples_per_second": 4.939,
|
| 13 |
+
"train_steps_per_second": 0.038
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 99.99,
|
| 3 |
+
"eval_loss": 0.5258437395095825,
|
| 4 |
+
"eval_runtime": 280.7122,
|
| 5 |
+
"eval_samples": 2693,
|
| 6 |
+
"eval_samples_per_second": 9.593,
|
| 7 |
+
"eval_steps_per_second": 0.602,
|
| 8 |
+
"eval_wer": 1.0073082607981607
|
| 9 |
+
}
|
runs/Feb05_21-52-35_0127-005726-ngyzqsk7-10-39-71-68/events.out.tfevents.1644179998.0127-005726-ngyzqsk7-10-39-71-68.8894.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:996c941caf925df4a35d2fd94291e934bd4cff9b194cc9a0c6be862eccfcbbbd
|
| 3 |
+
size 358
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 99.99,
|
| 3 |
+
"train_loss": 3.0460862658100742,
|
| 4 |
+
"train_runtime": 81000.8052,
|
| 5 |
+
"train_samples": 4001,
|
| 6 |
+
"train_samples_per_second": 4.939,
|
| 7 |
+
"train_steps_per_second": 0.038
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 99.98804780876495,
|
| 5 |
+
"global_step": 3100,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 3.22,
|
| 12 |
+
"learning_rate": 3.7125e-06,
|
| 13 |
+
"loss": 17.9778,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 6.45,
|
| 18 |
+
"learning_rate": 7.4625e-06,
|
| 19 |
+
"loss": 10.1788,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 9.67,
|
| 24 |
+
"learning_rate": 1.1212499999999998e-05,
|
| 25 |
+
"loss": 7.1097,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 12.89,
|
| 30 |
+
"learning_rate": 1.49625e-05,
|
| 31 |
+
"loss": 5.9162,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 16.13,
|
| 36 |
+
"learning_rate": 1.8712499999999997e-05,
|
| 37 |
+
"loss": 4.917,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 16.13,
|
| 42 |
+
"eval_loss": 4.89633846282959,
|
| 43 |
+
"eval_runtime": 284.283,
|
| 44 |
+
"eval_samples_per_second": 9.473,
|
| 45 |
+
"eval_steps_per_second": 0.594,
|
| 46 |
+
"eval_wer": 1.0,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 19.35,
|
| 51 |
+
"learning_rate": 2.2462499999999997e-05,
|
| 52 |
+
"loss": 4.1291,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 22.57,
|
| 57 |
+
"learning_rate": 2.6212499999999997e-05,
|
| 58 |
+
"loss": 3.6844,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 25.8,
|
| 63 |
+
"learning_rate": 2.99625e-05,
|
| 64 |
+
"loss": 3.5213,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 29.03,
|
| 69 |
+
"learning_rate": 3.37125e-05,
|
| 70 |
+
"loss": 3.4601,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 32.25,
|
| 75 |
+
"learning_rate": 3.7462499999999996e-05,
|
| 76 |
+
"loss": 3.3585,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 32.25,
|
| 81 |
+
"eval_loss": 3.3068907260894775,
|
| 82 |
+
"eval_runtime": 283.3986,
|
| 83 |
+
"eval_samples_per_second": 9.503,
|
| 84 |
+
"eval_steps_per_second": 0.596,
|
| 85 |
+
"eval_wer": 1.0000410576449335,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 35.48,
|
| 90 |
+
"learning_rate": 4.12125e-05,
|
| 91 |
+
"loss": 3.2956,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 38.7,
|
| 96 |
+
"learning_rate": 4.4962499999999995e-05,
|
| 97 |
+
"loss": 2.8135,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 41.92,
|
| 102 |
+
"learning_rate": 4.871249999999999e-05,
|
| 103 |
+
"loss": 2.0147,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 45.16,
|
| 108 |
+
"learning_rate": 5.2424999999999994e-05,
|
| 109 |
+
"loss": 1.7332,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 48.38,
|
| 114 |
+
"learning_rate": 5.6175e-05,
|
| 115 |
+
"loss": 1.5873,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 48.38,
|
| 120 |
+
"eval_loss": 0.8273974061012268,
|
| 121 |
+
"eval_runtime": 283.1526,
|
| 122 |
+
"eval_samples_per_second": 9.511,
|
| 123 |
+
"eval_steps_per_second": 0.597,
|
| 124 |
+
"eval_wer": 1.006076531450156,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 51.61,
|
| 129 |
+
"learning_rate": 5.9925e-05,
|
| 130 |
+
"loss": 1.476,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 54.83,
|
| 135 |
+
"learning_rate": 6.367499999999999e-05,
|
| 136 |
+
"loss": 1.4204,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 58.06,
|
| 141 |
+
"learning_rate": 6.7425e-05,
|
| 142 |
+
"loss": 1.357,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 61.29,
|
| 147 |
+
"learning_rate": 7.11375e-05,
|
| 148 |
+
"loss": 1.3037,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 64.51,
|
| 153 |
+
"learning_rate": 7.48875e-05,
|
| 154 |
+
"loss": 1.2654,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 64.51,
|
| 159 |
+
"eval_loss": 0.6249768137931824,
|
| 160 |
+
"eval_runtime": 282.7219,
|
| 161 |
+
"eval_samples_per_second": 9.525,
|
| 162 |
+
"eval_steps_per_second": 0.598,
|
| 163 |
+
"eval_wer": 1.0075546066677616,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 67.73,
|
| 168 |
+
"learning_rate": 6.838636363636363e-05,
|
| 169 |
+
"loss": 1.2231,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 70.96,
|
| 174 |
+
"learning_rate": 6.156818181818181e-05,
|
| 175 |
+
"loss": 1.1802,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 74.19,
|
| 180 |
+
"learning_rate": 5.481818181818182e-05,
|
| 181 |
+
"loss": 1.1484,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 77.41,
|
| 186 |
+
"learning_rate": 4.7999999999999994e-05,
|
| 187 |
+
"loss": 1.1185,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 80.64,
|
| 192 |
+
"learning_rate": 4.118181818181818e-05,
|
| 193 |
+
"loss": 1.0917,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 80.64,
|
| 198 |
+
"eval_loss": 0.5459514856338501,
|
| 199 |
+
"eval_runtime": 282.9764,
|
| 200 |
+
"eval_samples_per_second": 9.517,
|
| 201 |
+
"eval_steps_per_second": 0.597,
|
| 202 |
+
"eval_wer": 1.0056248973558877,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 83.86,
|
| 207 |
+
"learning_rate": 3.436363636363636e-05,
|
| 208 |
+
"loss": 1.0663,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 87.1,
|
| 213 |
+
"learning_rate": 2.754545454545454e-05,
|
| 214 |
+
"loss": 1.0487,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 90.32,
|
| 219 |
+
"learning_rate": 2.0727272727272725e-05,
|
| 220 |
+
"loss": 1.0214,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 93.54,
|
| 225 |
+
"learning_rate": 1.3909090909090907e-05,
|
| 226 |
+
"loss": 1.016,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 96.76,
|
| 231 |
+
"learning_rate": 7.09090909090909e-06,
|
| 232 |
+
"loss": 1.0001,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 96.76,
|
| 237 |
+
"eval_loss": 0.5304259061813354,
|
| 238 |
+
"eval_runtime": 282.7392,
|
| 239 |
+
"eval_samples_per_second": 9.525,
|
| 240 |
+
"eval_steps_per_second": 0.598,
|
| 241 |
+
"eval_wer": 1.0082525866316308,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 99.99,
|
| 246 |
+
"learning_rate": 2.727272727272727e-07,
|
| 247 |
+
"loss": 0.9947,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 99.99,
|
| 252 |
+
"step": 3100,
|
| 253 |
+
"total_flos": 5.0567037565307716e+19,
|
| 254 |
+
"train_loss": 3.0460862658100742,
|
| 255 |
+
"train_runtime": 81000.8052,
|
| 256 |
+
"train_samples_per_second": 4.939,
|
| 257 |
+
"train_steps_per_second": 0.038
|
| 258 |
+
}
|
| 259 |
+
],
|
| 260 |
+
"max_steps": 3100,
|
| 261 |
+
"num_train_epochs": 100,
|
| 262 |
+
"total_flos": 5.0567037565307716e+19,
|
| 263 |
+
"trial_name": null,
|
| 264 |
+
"trial_params": null
|
| 265 |
+
}
|