End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +220 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.35227298736572266,
|
| 4 |
+
"eval_runtime": 296.525,
|
| 5 |
+
"eval_samples": 5564,
|
| 6 |
+
"eval_samples_per_second": 18.764,
|
| 7 |
+
"eval_steps_per_second": 4.691,
|
| 8 |
+
"eval_wer": 0.20457435984398847,
|
| 9 |
+
"train_loss": 0.3001849569908102,
|
| 10 |
+
"train_runtime": 25737.7904,
|
| 11 |
+
"train_samples": 14316,
|
| 12 |
+
"train_samples_per_second": 8.343,
|
| 13 |
+
"train_steps_per_second": 0.261
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.35227298736572266,
|
| 4 |
+
"eval_runtime": 296.525,
|
| 5 |
+
"eval_samples": 5564,
|
| 6 |
+
"eval_samples_per_second": 18.764,
|
| 7 |
+
"eval_steps_per_second": 4.691,
|
| 8 |
+
"eval_wer": 0.20457435984398847
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 0.3001849569908102,
|
| 4 |
+
"train_runtime": 25737.7904,
|
| 5 |
+
"train_samples": 14316,
|
| 6 |
+
"train_samples_per_second": 8.343,
|
| 7 |
+
"train_steps_per_second": 0.261
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,220 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 14.999161777032691,
|
| 5 |
+
"global_step": 6705,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.12,
|
| 12 |
+
"learning_rate": 0.00029939999999999996,
|
| 13 |
+
"loss": 3.0536,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.12,
|
| 18 |
+
"eval_loss": 0.5348725914955139,
|
| 19 |
+
"eval_runtime": 293.4912,
|
| 20 |
+
"eval_samples_per_second": 18.958,
|
| 21 |
+
"eval_steps_per_second": 4.739,
|
| 22 |
+
"eval_wer": 0.43377988807868406,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 2.24,
|
| 27 |
+
"learning_rate": 0.0002758742949234488,
|
| 28 |
+
"loss": 0.2543,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 2.24,
|
| 33 |
+
"eval_loss": 0.38594746589660645,
|
| 34 |
+
"eval_runtime": 294.551,
|
| 35 |
+
"eval_samples_per_second": 18.89,
|
| 36 |
+
"eval_steps_per_second": 4.722,
|
| 37 |
+
"eval_wer": 0.3029082584364931,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 3.36,
|
| 42 |
+
"learning_rate": 0.0002517002417405318,
|
| 43 |
+
"loss": 0.1472,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 3.36,
|
| 48 |
+
"eval_loss": 0.34706196188926697,
|
| 49 |
+
"eval_runtime": 295.2527,
|
| 50 |
+
"eval_samples_per_second": 18.845,
|
| 51 |
+
"eval_steps_per_second": 4.711,
|
| 52 |
+
"eval_wer": 0.28175343394946584,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 4.47,
|
| 57 |
+
"learning_rate": 0.00022752618855761478,
|
| 58 |
+
"loss": 0.1088,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 4.47,
|
| 63 |
+
"eval_loss": 0.34891438484191895,
|
| 64 |
+
"eval_runtime": 293.508,
|
| 65 |
+
"eval_samples_per_second": 18.957,
|
| 66 |
+
"eval_steps_per_second": 4.739,
|
| 67 |
+
"eval_wer": 0.27306257419026625,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.59,
|
| 72 |
+
"learning_rate": 0.00020340048348106365,
|
| 73 |
+
"loss": 0.0855,
|
| 74 |
+
"step": 2500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 5.59,
|
| 78 |
+
"eval_loss": 0.3582000732421875,
|
| 79 |
+
"eval_runtime": 293.4793,
|
| 80 |
+
"eval_samples_per_second": 18.959,
|
| 81 |
+
"eval_steps_per_second": 4.74,
|
| 82 |
+
"eval_wer": 0.25578684076649144,
|
| 83 |
+
"step": 2500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 6.71,
|
| 87 |
+
"learning_rate": 0.00017922643029814666,
|
| 88 |
+
"loss": 0.0721,
|
| 89 |
+
"step": 3000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.71,
|
| 93 |
+
"eval_loss": 0.34574663639068604,
|
| 94 |
+
"eval_runtime": 293.1368,
|
| 95 |
+
"eval_samples_per_second": 18.981,
|
| 96 |
+
"eval_steps_per_second": 4.745,
|
| 97 |
+
"eval_wer": 0.24711717822621673,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.83,
|
| 102 |
+
"learning_rate": 0.00015505237711522965,
|
| 103 |
+
"loss": 0.0653,
|
| 104 |
+
"step": 3500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 7.83,
|
| 108 |
+
"eval_loss": 0.3298647999763489,
|
| 109 |
+
"eval_runtime": 293.814,
|
| 110 |
+
"eval_samples_per_second": 18.937,
|
| 111 |
+
"eval_steps_per_second": 4.734,
|
| 112 |
+
"eval_wer": 0.23569187722570797,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 8.95,
|
| 117 |
+
"learning_rate": 0.00013087832393231263,
|
| 118 |
+
"loss": 0.0527,
|
| 119 |
+
"step": 4000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 8.95,
|
| 123 |
+
"eval_loss": 0.3439774811267853,
|
| 124 |
+
"eval_runtime": 293.0867,
|
| 125 |
+
"eval_samples_per_second": 18.984,
|
| 126 |
+
"eval_steps_per_second": 4.746,
|
| 127 |
+
"eval_wer": 0.23338138036289638,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 10.07,
|
| 132 |
+
"learning_rate": 0.00010670427074939564,
|
| 133 |
+
"loss": 0.0444,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.07,
|
| 138 |
+
"eval_loss": 0.34170448780059814,
|
| 139 |
+
"eval_runtime": 294.6348,
|
| 140 |
+
"eval_samples_per_second": 18.884,
|
| 141 |
+
"eval_steps_per_second": 4.721,
|
| 142 |
+
"eval_wer": 0.22890876716974734,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 11.19,
|
| 147 |
+
"learning_rate": 8.253021756647865e-05,
|
| 148 |
+
"loss": 0.0404,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 11.19,
|
| 153 |
+
"eval_loss": 0.36909687519073486,
|
| 154 |
+
"eval_runtime": 294.7932,
|
| 155 |
+
"eval_samples_per_second": 18.874,
|
| 156 |
+
"eval_steps_per_second": 4.719,
|
| 157 |
+
"eval_wer": 0.2204298795997965,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 12.3,
|
| 162 |
+
"learning_rate": 5.835616438356164e-05,
|
| 163 |
+
"loss": 0.0345,
|
| 164 |
+
"step": 5500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 12.3,
|
| 168 |
+
"eval_loss": 0.3453255593776703,
|
| 169 |
+
"eval_runtime": 293.4104,
|
| 170 |
+
"eval_samples_per_second": 18.963,
|
| 171 |
+
"eval_steps_per_second": 4.741,
|
| 172 |
+
"eval_wer": 0.21021282007800576,
|
| 173 |
+
"step": 5500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 13.42,
|
| 177 |
+
"learning_rate": 3.418211120064464e-05,
|
| 178 |
+
"loss": 0.0288,
|
| 179 |
+
"step": 6000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 13.42,
|
| 183 |
+
"eval_loss": 0.3634118437767029,
|
| 184 |
+
"eval_runtime": 296.1491,
|
| 185 |
+
"eval_samples_per_second": 18.788,
|
| 186 |
+
"eval_steps_per_second": 4.697,
|
| 187 |
+
"eval_wer": 0.20885619806681363,
|
| 188 |
+
"step": 6000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 14.54,
|
| 192 |
+
"learning_rate": 1.0008058017727637e-05,
|
| 193 |
+
"loss": 0.027,
|
| 194 |
+
"step": 6500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 14.54,
|
| 198 |
+
"eval_loss": 0.35320568084716797,
|
| 199 |
+
"eval_runtime": 290.8751,
|
| 200 |
+
"eval_samples_per_second": 19.128,
|
| 201 |
+
"eval_steps_per_second": 4.782,
|
| 202 |
+
"eval_wer": 0.20438358487366456,
|
| 203 |
+
"step": 6500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 15.0,
|
| 207 |
+
"step": 6705,
|
| 208 |
+
"total_flos": 2.897461391875666e+19,
|
| 209 |
+
"train_loss": 0.3001849569908102,
|
| 210 |
+
"train_runtime": 25737.7904,
|
| 211 |
+
"train_samples_per_second": 8.343,
|
| 212 |
+
"train_steps_per_second": 0.261
|
| 213 |
+
}
|
| 214 |
+
],
|
| 215 |
+
"max_steps": 6705,
|
| 216 |
+
"num_train_epochs": 15,
|
| 217 |
+
"total_flos": 2.897461391875666e+19,
|
| 218 |
+
"trial_name": null,
|
| 219 |
+
"trial_params": null
|
| 220 |
+
}
|