End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.86,
|
| 3 |
+
"eval_loss": 1.019495964050293,
|
| 4 |
+
"eval_runtime": 253.9493,
|
| 5 |
+
"eval_samples_per_second": 2.016,
|
| 6 |
+
"eval_steps_per_second": 0.126,
|
| 7 |
+
"eval_wer": 66.14709443099274,
|
| 8 |
+
"train_loss": 0.6091492390632629,
|
| 9 |
+
"train_runtime": 1414.6972,
|
| 10 |
+
"train_samples_per_second": 13.572,
|
| 11 |
+
"train_steps_per_second": 0.212
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.86,
|
| 3 |
+
"eval_loss": 1.019495964050293,
|
| 4 |
+
"eval_runtime": 253.9493,
|
| 5 |
+
"eval_samples_per_second": 2.016,
|
| 6 |
+
"eval_steps_per_second": 0.126,
|
| 7 |
+
"eval_wer": 66.14709443099274
|
| 8 |
+
}
|
runs/Dec18_15-04-30_129-146-179-188/events.out.tfevents.1671377637.129-146-179-188.129110.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d8af1e4495e453d3178ce9f999e74f09dabce53ec8196c68e996a1f080bfb5a7
|
| 3 |
+
size 358
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.86,
|
| 3 |
+
"train_loss": 0.6091492390632629,
|
| 4 |
+
"train_runtime": 1414.6972,
|
| 5 |
+
"train_samples_per_second": 13.572,
|
| 6 |
+
"train_steps_per_second": 0.212
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,232 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.019495964050293,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-100",
|
| 4 |
+
"epoch": 42.857142857142854,
|
| 5 |
+
"global_step": 300,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.43,
|
| 12 |
+
"learning_rate": 1.75e-06,
|
| 13 |
+
"loss": 3.252,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 2.86,
|
| 18 |
+
"learning_rate": 4.25e-06,
|
| 19 |
+
"loss": 2.8169,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 4.29,
|
| 24 |
+
"learning_rate": 6.750000000000001e-06,
|
| 25 |
+
"loss": 2.2237,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 5.71,
|
| 30 |
+
"learning_rate": 9.250000000000001e-06,
|
| 31 |
+
"loss": 1.6944,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 7.14,
|
| 36 |
+
"learning_rate": 9.730769230769231e-06,
|
| 37 |
+
"loss": 1.2764,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 8.57,
|
| 42 |
+
"learning_rate": 9.346153846153847e-06,
|
| 43 |
+
"loss": 1.0164,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 10.0,
|
| 48 |
+
"learning_rate": 8.961538461538462e-06,
|
| 49 |
+
"loss": 0.854,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 11.43,
|
| 54 |
+
"learning_rate": 8.576923076923077e-06,
|
| 55 |
+
"loss": 0.7109,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 12.86,
|
| 60 |
+
"learning_rate": 8.192307692307692e-06,
|
| 61 |
+
"loss": 0.613,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 14.29,
|
| 66 |
+
"learning_rate": 7.807692307692309e-06,
|
| 67 |
+
"loss": 0.5307,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 14.29,
|
| 72 |
+
"eval_loss": 1.019495964050293,
|
| 73 |
+
"eval_runtime": 229.2012,
|
| 74 |
+
"eval_samples_per_second": 2.234,
|
| 75 |
+
"eval_steps_per_second": 0.14,
|
| 76 |
+
"eval_wer": 66.14709443099274,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 15.71,
|
| 81 |
+
"learning_rate": 7.423076923076924e-06,
|
| 82 |
+
"loss": 0.457,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 17.14,
|
| 87 |
+
"learning_rate": 7.038461538461539e-06,
|
| 88 |
+
"loss": 0.3915,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 18.57,
|
| 93 |
+
"learning_rate": 6.653846153846154e-06,
|
| 94 |
+
"loss": 0.3438,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 20.0,
|
| 99 |
+
"learning_rate": 6.26923076923077e-06,
|
| 100 |
+
"loss": 0.2985,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 21.43,
|
| 105 |
+
"learning_rate": 5.884615384615385e-06,
|
| 106 |
+
"loss": 0.2591,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 22.86,
|
| 111 |
+
"learning_rate": 5.500000000000001e-06,
|
| 112 |
+
"loss": 0.2199,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 24.29,
|
| 117 |
+
"learning_rate": 5.115384615384616e-06,
|
| 118 |
+
"loss": 0.1928,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 25.71,
|
| 123 |
+
"learning_rate": 4.730769230769231e-06,
|
| 124 |
+
"loss": 0.1676,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 27.14,
|
| 129 |
+
"learning_rate": 4.346153846153846e-06,
|
| 130 |
+
"loss": 0.1435,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 28.57,
|
| 135 |
+
"learning_rate": 3.961538461538462e-06,
|
| 136 |
+
"loss": 0.1225,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 28.57,
|
| 141 |
+
"eval_loss": 1.146507740020752,
|
| 142 |
+
"eval_runtime": 276.2984,
|
| 143 |
+
"eval_samples_per_second": 1.853,
|
| 144 |
+
"eval_steps_per_second": 0.116,
|
| 145 |
+
"eval_wer": 66.01846246973365,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 30.0,
|
| 150 |
+
"learning_rate": 3.5769230769230773e-06,
|
| 151 |
+
"loss": 0.108,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 31.43,
|
| 156 |
+
"learning_rate": 3.192307692307692e-06,
|
| 157 |
+
"loss": 0.0941,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 32.86,
|
| 162 |
+
"learning_rate": 2.807692307692308e-06,
|
| 163 |
+
"loss": 0.0811,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 34.29,
|
| 168 |
+
"learning_rate": 2.4230769230769233e-06,
|
| 169 |
+
"loss": 0.0729,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 35.71,
|
| 174 |
+
"learning_rate": 2.0384615384615386e-06,
|
| 175 |
+
"loss": 0.0655,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 37.14,
|
| 180 |
+
"learning_rate": 1.653846153846154e-06,
|
| 181 |
+
"loss": 0.0604,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 38.57,
|
| 186 |
+
"learning_rate": 1.2692307692307692e-06,
|
| 187 |
+
"loss": 0.055,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 40.0,
|
| 192 |
+
"learning_rate": 8.846153846153848e-07,
|
| 193 |
+
"loss": 0.0524,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 41.43,
|
| 198 |
+
"learning_rate": 5.000000000000001e-07,
|
| 199 |
+
"loss": 0.0505,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 42.86,
|
| 204 |
+
"learning_rate": 1.153846153846154e-07,
|
| 205 |
+
"loss": 0.0498,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 42.86,
|
| 210 |
+
"eval_loss": 1.229990839958191,
|
| 211 |
+
"eval_runtime": 273.4631,
|
| 212 |
+
"eval_samples_per_second": 1.872,
|
| 213 |
+
"eval_steps_per_second": 0.117,
|
| 214 |
+
"eval_wer": 66.03359564164649,
|
| 215 |
+
"step": 300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 42.86,
|
| 219 |
+
"step": 300,
|
| 220 |
+
"total_flos": 1.19900188901376e+18,
|
| 221 |
+
"train_loss": 0.6091492390632629,
|
| 222 |
+
"train_runtime": 1414.6972,
|
| 223 |
+
"train_samples_per_second": 13.572,
|
| 224 |
+
"train_steps_per_second": 0.212
|
| 225 |
+
}
|
| 226 |
+
],
|
| 227 |
+
"max_steps": 300,
|
| 228 |
+
"num_train_epochs": 43,
|
| 229 |
+
"total_flos": 1.19900188901376e+18,
|
| 230 |
+
"trial_name": null,
|
| 231 |
+
"trial_params": null
|
| 232 |
+
}
|