Commit ·
464fd90
1
Parent(s): 86fbc5e
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- runs/May24_15-42-26_gpu3060/events.out.tfevents.1685013355.gpu3060.913331.2 +3 -0
- train_results.json +8 -0
- trainer_state.json +520 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.04323260486125946,
|
| 4 |
+
"eval_runtime": 1030.4809,
|
| 5 |
+
"eval_samples": 13999,
|
| 6 |
+
"eval_samples_per_second": 13.585,
|
| 7 |
+
"eval_steps_per_second": 1.698,
|
| 8 |
+
"eval_wer": 0.03220236790177593,
|
| 9 |
+
"train_loss": 0.18964702577639164,
|
| 10 |
+
"train_runtime": 68991.5968,
|
| 11 |
+
"train_samples": 23999,
|
| 12 |
+
"train_samples_per_second": 5.218,
|
| 13 |
+
"train_steps_per_second": 0.373
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"eval_loss": 0.04323260486125946,
|
| 4 |
+
"eval_runtime": 1030.4809,
|
| 5 |
+
"eval_samples": 13999,
|
| 6 |
+
"eval_samples_per_second": 13.585,
|
| 7 |
+
"eval_steps_per_second": 1.698,
|
| 8 |
+
"eval_wer": 0.03220236790177593
|
| 9 |
+
}
|
runs/May24_15-42-26_gpu3060/events.out.tfevents.1685013355.gpu3060.913331.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fb9c2dd173e5de954641abe13ec55c70f0609d7d503eba9f631030576451538
|
| 3 |
+
size 364
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"train_loss": 0.18964702577639164,
|
| 4 |
+
"train_runtime": 68991.5968,
|
| 5 |
+
"train_samples": 23999,
|
| 6 |
+
"train_samples_per_second": 5.218,
|
| 7 |
+
"train_steps_per_second": 0.373
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,520 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 14.995625546806648,
|
| 5 |
+
"global_step": 25710,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.29,
|
| 12 |
+
"learning_rate": 0.0002988,
|
| 13 |
+
"loss": 3.7959,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.58,
|
| 18 |
+
"learning_rate": 0.00029407378024593416,
|
| 19 |
+
"loss": 0.4885,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.7,
|
| 24 |
+
"eval_loss": 0.29581278562545776,
|
| 25 |
+
"eval_runtime": 1055.6593,
|
| 26 |
+
"eval_samples_per_second": 13.261,
|
| 27 |
+
"eval_steps_per_second": 1.658,
|
| 28 |
+
"eval_wer": 0.26183950887963164,
|
| 29 |
+
"step": 1200
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.87,
|
| 33 |
+
"learning_rate": 0.0002881237604125347,
|
| 34 |
+
"loss": 0.3691,
|
| 35 |
+
"step": 1500
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.17,
|
| 39 |
+
"learning_rate": 0.0002821737405791352,
|
| 40 |
+
"loss": 0.2986,
|
| 41 |
+
"step": 2000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.4,
|
| 45 |
+
"eval_loss": 0.18017804622650146,
|
| 46 |
+
"eval_runtime": 1040.4563,
|
| 47 |
+
"eval_samples_per_second": 13.455,
|
| 48 |
+
"eval_steps_per_second": 1.682,
|
| 49 |
+
"eval_wer": 0.16287546590659943,
|
| 50 |
+
"step": 2400
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.46,
|
| 54 |
+
"learning_rate": 0.0002762237207457358,
|
| 55 |
+
"loss": 0.2714,
|
| 56 |
+
"step": 2500
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.75,
|
| 60 |
+
"learning_rate": 0.00027027370091233633,
|
| 61 |
+
"loss": 0.2631,
|
| 62 |
+
"step": 3000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.04,
|
| 66 |
+
"learning_rate": 0.0002643236810789369,
|
| 67 |
+
"loss": 0.2515,
|
| 68 |
+
"step": 3500
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.1,
|
| 72 |
+
"eval_loss": 0.13786812126636505,
|
| 73 |
+
"eval_runtime": 1014.0923,
|
| 74 |
+
"eval_samples_per_second": 13.804,
|
| 75 |
+
"eval_steps_per_second": 1.726,
|
| 76 |
+
"eval_wer": 0.13172823942117956,
|
| 77 |
+
"step": 3600
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2.33,
|
| 81 |
+
"learning_rate": 0.0002583736612455375,
|
| 82 |
+
"loss": 0.2002,
|
| 83 |
+
"step": 4000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.62,
|
| 87 |
+
"learning_rate": 0.000252423641412138,
|
| 88 |
+
"loss": 0.2013,
|
| 89 |
+
"step": 4500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 2.8,
|
| 93 |
+
"eval_loss": 0.12080267816781998,
|
| 94 |
+
"eval_runtime": 1022.088,
|
| 95 |
+
"eval_samples_per_second": 13.696,
|
| 96 |
+
"eval_steps_per_second": 1.712,
|
| 97 |
+
"eval_wer": 0.11781955711466784,
|
| 98 |
+
"step": 4800
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.92,
|
| 102 |
+
"learning_rate": 0.0002464736215787386,
|
| 103 |
+
"loss": 0.1999,
|
| 104 |
+
"step": 5000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 3.21,
|
| 108 |
+
"learning_rate": 0.00024052360174533913,
|
| 109 |
+
"loss": 0.172,
|
| 110 |
+
"step": 5500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 3.5,
|
| 114 |
+
"learning_rate": 0.0002345735819119397,
|
| 115 |
+
"loss": 0.1651,
|
| 116 |
+
"step": 6000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 3.5,
|
| 120 |
+
"eval_loss": 0.11095409095287323,
|
| 121 |
+
"eval_runtime": 1036.2931,
|
| 122 |
+
"eval_samples_per_second": 13.509,
|
| 123 |
+
"eval_steps_per_second": 1.689,
|
| 124 |
+
"eval_wer": 0.11590796974347731,
|
| 125 |
+
"step": 6000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.79,
|
| 129 |
+
"learning_rate": 0.00022862356207854024,
|
| 130 |
+
"loss": 0.1642,
|
| 131 |
+
"step": 6500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 4.08,
|
| 135 |
+
"learning_rate": 0.00022267354224514083,
|
| 136 |
+
"loss": 0.1559,
|
| 137 |
+
"step": 7000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 4.2,
|
| 141 |
+
"eval_loss": 0.09234469383955002,
|
| 142 |
+
"eval_runtime": 1044.1963,
|
| 143 |
+
"eval_samples_per_second": 13.406,
|
| 144 |
+
"eval_steps_per_second": 1.676,
|
| 145 |
+
"eval_wer": 0.09475032887524666,
|
| 146 |
+
"step": 7200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 4.37,
|
| 150 |
+
"learning_rate": 0.00021672352241174135,
|
| 151 |
+
"loss": 0.1407,
|
| 152 |
+
"step": 7500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.67,
|
| 156 |
+
"learning_rate": 0.0002107854026180087,
|
| 157 |
+
"loss": 0.1337,
|
| 158 |
+
"step": 8000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 4.9,
|
| 162 |
+
"eval_loss": 0.09278657287359238,
|
| 163 |
+
"eval_runtime": 1042.5487,
|
| 164 |
+
"eval_samples_per_second": 13.428,
|
| 165 |
+
"eval_steps_per_second": 1.679,
|
| 166 |
+
"eval_wer": 0.09305114010085508,
|
| 167 |
+
"step": 8400
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 4.96,
|
| 171 |
+
"learning_rate": 0.00020483538278460924,
|
| 172 |
+
"loss": 0.1391,
|
| 173 |
+
"step": 8500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 5.25,
|
| 177 |
+
"learning_rate": 0.00019889726299087662,
|
| 178 |
+
"loss": 0.1241,
|
| 179 |
+
"step": 9000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 5.54,
|
| 183 |
+
"learning_rate": 0.00019294724315747718,
|
| 184 |
+
"loss": 0.1162,
|
| 185 |
+
"step": 9500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 5.6,
|
| 189 |
+
"eval_loss": 0.07527832686901093,
|
| 190 |
+
"eval_runtime": 1045.0498,
|
| 191 |
+
"eval_samples_per_second": 13.396,
|
| 192 |
+
"eval_steps_per_second": 1.675,
|
| 193 |
+
"eval_wer": 0.07819008989256741,
|
| 194 |
+
"step": 9600
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.83,
|
| 198 |
+
"learning_rate": 0.0001869972233240777,
|
| 199 |
+
"loss": 0.1203,
|
| 200 |
+
"step": 10000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 6.12,
|
| 204 |
+
"learning_rate": 0.0001810472034906783,
|
| 205 |
+
"loss": 0.1164,
|
| 206 |
+
"step": 10500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 6.3,
|
| 210 |
+
"eval_loss": 0.07002939283847809,
|
| 211 |
+
"eval_runtime": 1036.4649,
|
| 212 |
+
"eval_samples_per_second": 13.506,
|
| 213 |
+
"eval_steps_per_second": 1.688,
|
| 214 |
+
"eval_wer": 0.07144129576847183,
|
| 215 |
+
"step": 10800
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 6.42,
|
| 219 |
+
"learning_rate": 0.00017509718365727885,
|
| 220 |
+
"loss": 0.1023,
|
| 221 |
+
"step": 11000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 6.71,
|
| 225 |
+
"learning_rate": 0.00016914716382387937,
|
| 226 |
+
"loss": 0.1034,
|
| 227 |
+
"step": 11500
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 7.0,
|
| 231 |
+
"learning_rate": 0.00016319714399047996,
|
| 232 |
+
"loss": 0.1057,
|
| 233 |
+
"step": 12000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 7.0,
|
| 237 |
+
"eval_loss": 0.06297972798347473,
|
| 238 |
+
"eval_runtime": 1044.9779,
|
| 239 |
+
"eval_samples_per_second": 13.396,
|
| 240 |
+
"eval_steps_per_second": 1.675,
|
| 241 |
+
"eval_wer": 0.06563801797851349,
|
| 242 |
+
"step": 12000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 7.29,
|
| 246 |
+
"learning_rate": 0.0001572471241570805,
|
| 247 |
+
"loss": 0.085,
|
| 248 |
+
"step": 12500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 7.58,
|
| 252 |
+
"learning_rate": 0.00015129710432368104,
|
| 253 |
+
"loss": 0.0904,
|
| 254 |
+
"step": 13000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 7.7,
|
| 258 |
+
"eval_loss": 0.061889298260211945,
|
| 259 |
+
"eval_runtime": 1047.4068,
|
| 260 |
+
"eval_samples_per_second": 13.365,
|
| 261 |
+
"eval_steps_per_second": 1.671,
|
| 262 |
+
"eval_wer": 0.06239037491778119,
|
| 263 |
+
"step": 13200
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 7.87,
|
| 267 |
+
"learning_rate": 0.00014534708449028162,
|
| 268 |
+
"loss": 0.0893,
|
| 269 |
+
"step": 13500
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 8.17,
|
| 273 |
+
"learning_rate": 0.00013939706465688218,
|
| 274 |
+
"loss": 0.0807,
|
| 275 |
+
"step": 14000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 8.4,
|
| 279 |
+
"eval_loss": 0.06088259816169739,
|
| 280 |
+
"eval_runtime": 1049.8382,
|
| 281 |
+
"eval_samples_per_second": 13.334,
|
| 282 |
+
"eval_steps_per_second": 1.667,
|
| 283 |
+
"eval_wer": 0.05664876123657093,
|
| 284 |
+
"step": 14400
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 8.46,
|
| 288 |
+
"learning_rate": 0.00013345894486314954,
|
| 289 |
+
"loss": 0.0778,
|
| 290 |
+
"step": 14500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 8.75,
|
| 294 |
+
"learning_rate": 0.0001275089250297501,
|
| 295 |
+
"loss": 0.0752,
|
| 296 |
+
"step": 15000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 9.04,
|
| 300 |
+
"learning_rate": 0.00012157080523601744,
|
| 301 |
+
"loss": 0.0759,
|
| 302 |
+
"step": 15500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 9.1,
|
| 306 |
+
"eval_loss": 0.05135694146156311,
|
| 307 |
+
"eval_runtime": 1026.2376,
|
| 308 |
+
"eval_samples_per_second": 13.641,
|
| 309 |
+
"eval_steps_per_second": 1.705,
|
| 310 |
+
"eval_wer": 0.04897500548125411,
|
| 311 |
+
"step": 15600
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 9.33,
|
| 315 |
+
"learning_rate": 0.000115620785402618,
|
| 316 |
+
"loss": 0.0681,
|
| 317 |
+
"step": 16000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 9.62,
|
| 321 |
+
"learning_rate": 0.00010967076556921856,
|
| 322 |
+
"loss": 0.0657,
|
| 323 |
+
"step": 16500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 9.8,
|
| 327 |
+
"eval_loss": 0.05039794743061066,
|
| 328 |
+
"eval_runtime": 1044.7693,
|
| 329 |
+
"eval_samples_per_second": 13.399,
|
| 330 |
+
"eval_steps_per_second": 1.675,
|
| 331 |
+
"eval_wer": 0.04700860556895418,
|
| 332 |
+
"step": 16800
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 9.92,
|
| 336 |
+
"learning_rate": 0.0001037207457358191,
|
| 337 |
+
"loss": 0.0672,
|
| 338 |
+
"step": 17000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 10.21,
|
| 342 |
+
"learning_rate": 9.777072590241967e-05,
|
| 343 |
+
"loss": 0.0597,
|
| 344 |
+
"step": 17500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 10.5,
|
| 348 |
+
"learning_rate": 9.183260610868703e-05,
|
| 349 |
+
"loss": 0.0556,
|
| 350 |
+
"step": 18000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 10.5,
|
| 354 |
+
"eval_loss": 0.05111551284790039,
|
| 355 |
+
"eval_runtime": 1048.1552,
|
| 356 |
+
"eval_samples_per_second": 13.356,
|
| 357 |
+
"eval_steps_per_second": 1.67,
|
| 358 |
+
"eval_wer": 0.04308950887963166,
|
| 359 |
+
"step": 18000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 10.79,
|
| 363 |
+
"learning_rate": 8.588258627528757e-05,
|
| 364 |
+
"loss": 0.0581,
|
| 365 |
+
"step": 18500
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 11.08,
|
| 369 |
+
"learning_rate": 7.993256644188814e-05,
|
| 370 |
+
"loss": 0.0534,
|
| 371 |
+
"step": 19000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 11.2,
|
| 375 |
+
"eval_loss": 0.04835359379649162,
|
| 376 |
+
"eval_runtime": 1051.2322,
|
| 377 |
+
"eval_samples_per_second": 13.317,
|
| 378 |
+
"eval_steps_per_second": 1.665,
|
| 379 |
+
"eval_wer": 0.040807936855952645,
|
| 380 |
+
"step": 19200
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 11.37,
|
| 384 |
+
"learning_rate": 7.39825466084887e-05,
|
| 385 |
+
"loss": 0.0537,
|
| 386 |
+
"step": 19500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 11.67,
|
| 390 |
+
"learning_rate": 6.803252677508924e-05,
|
| 391 |
+
"loss": 0.0498,
|
| 392 |
+
"step": 20000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 11.9,
|
| 396 |
+
"eval_loss": 0.04363071173429489,
|
| 397 |
+
"eval_runtime": 1040.6039,
|
| 398 |
+
"eval_samples_per_second": 13.453,
|
| 399 |
+
"eval_steps_per_second": 1.682,
|
| 400 |
+
"eval_wer": 0.03834822407366806,
|
| 401 |
+
"step": 20400
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 11.96,
|
| 405 |
+
"learning_rate": 6.20825069416898e-05,
|
| 406 |
+
"loss": 0.0493,
|
| 407 |
+
"step": 20500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 12.25,
|
| 411 |
+
"learning_rate": 5.613248710829036e-05,
|
| 412 |
+
"loss": 0.047,
|
| 413 |
+
"step": 21000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 12.54,
|
| 417 |
+
"learning_rate": 5.018246727489091e-05,
|
| 418 |
+
"loss": 0.0441,
|
| 419 |
+
"step": 21500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 12.6,
|
| 423 |
+
"eval_loss": 0.04579892382025719,
|
| 424 |
+
"eval_runtime": 1085.7641,
|
| 425 |
+
"eval_samples_per_second": 12.893,
|
| 426 |
+
"eval_steps_per_second": 1.612,
|
| 427 |
+
"eval_wer": 0.03654626178469634,
|
| 428 |
+
"step": 21600
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 12.83,
|
| 432 |
+
"learning_rate": 4.423244744149147e-05,
|
| 433 |
+
"loss": 0.0407,
|
| 434 |
+
"step": 22000
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 13.12,
|
| 438 |
+
"learning_rate": 3.8282427608092026e-05,
|
| 439 |
+
"loss": 0.0398,
|
| 440 |
+
"step": 22500
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 13.3,
|
| 444 |
+
"eval_loss": 0.047090090811252594,
|
| 445 |
+
"eval_runtime": 1087.0658,
|
| 446 |
+
"eval_samples_per_second": 12.878,
|
| 447 |
+
"eval_steps_per_second": 1.61,
|
| 448 |
+
"eval_wer": 0.035360940583205436,
|
| 449 |
+
"step": 22800
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 13.41,
|
| 453 |
+
"learning_rate": 3.233240777469258e-05,
|
| 454 |
+
"loss": 0.0377,
|
| 455 |
+
"step": 23000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 13.71,
|
| 459 |
+
"learning_rate": 2.6382387941293133e-05,
|
| 460 |
+
"loss": 0.0372,
|
| 461 |
+
"step": 23500
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 14.0,
|
| 465 |
+
"learning_rate": 2.044426814756049e-05,
|
| 466 |
+
"loss": 0.0379,
|
| 467 |
+
"step": 24000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 14.0,
|
| 471 |
+
"eval_loss": 0.04015319421887398,
|
| 472 |
+
"eval_runtime": 1040.8936,
|
| 473 |
+
"eval_samples_per_second": 13.449,
|
| 474 |
+
"eval_steps_per_second": 1.681,
|
| 475 |
+
"eval_wer": 0.03266827450120587,
|
| 476 |
+
"step": 24000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 14.29,
|
| 480 |
+
"learning_rate": 1.4494248314161046e-05,
|
| 481 |
+
"loss": 0.0349,
|
| 482 |
+
"step": 24500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 14.58,
|
| 486 |
+
"learning_rate": 8.544228480761601e-06,
|
| 487 |
+
"loss": 0.0333,
|
| 488 |
+
"step": 25000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 14.7,
|
| 492 |
+
"eval_loss": 0.04384541139006615,
|
| 493 |
+
"eval_runtime": 1040.8091,
|
| 494 |
+
"eval_samples_per_second": 13.45,
|
| 495 |
+
"eval_steps_per_second": 1.681,
|
| 496 |
+
"eval_wer": 0.032620313527735145,
|
| 497 |
+
"step": 25200
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 14.87,
|
| 501 |
+
"learning_rate": 2.594208647362158e-06,
|
| 502 |
+
"loss": 0.0316,
|
| 503 |
+
"step": 25500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 15.0,
|
| 507 |
+
"step": 25710,
|
| 508 |
+
"total_flos": 5.133012941501205e+19,
|
| 509 |
+
"train_loss": 0.18964702577639164,
|
| 510 |
+
"train_runtime": 68991.5968,
|
| 511 |
+
"train_samples_per_second": 5.218,
|
| 512 |
+
"train_steps_per_second": 0.373
|
| 513 |
+
}
|
| 514 |
+
],
|
| 515 |
+
"max_steps": 25710,
|
| 516 |
+
"num_train_epochs": 15,
|
| 517 |
+
"total_flos": 5.133012941501205e+19,
|
| 518 |
+
"trial_name": null,
|
| 519 |
+
"trial_params": null
|
| 520 |
+
}
|