Commit
·
2e1cb14
1
Parent(s):
ad8bf35
End of training
Browse files- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +649 -0
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.0,
|
| 3 |
+
"eval_cer": 0.02656322252479063,
|
| 4 |
+
"eval_loss": 0.1773720234632492,
|
| 5 |
+
"eval_runtime": 20.504,
|
| 6 |
+
"eval_samples": 312,
|
| 7 |
+
"eval_samples_per_second": 15.217,
|
| 8 |
+
"eval_steps_per_second": 1.902,
|
| 9 |
+
"eval_wer": 0.08439275087909115,
|
| 10 |
+
"total_flos": 8.489873279367732e+19,
|
| 11 |
+
"train_loss": 0.24954385395291484,
|
| 12 |
+
"train_runtime": 14687.2421,
|
| 13 |
+
"train_samples": 2517,
|
| 14 |
+
"train_samples_per_second": 17.137,
|
| 15 |
+
"train_steps_per_second": 0.538
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.0,
|
| 3 |
+
"eval_cer": 0.02656322252479063,
|
| 4 |
+
"eval_loss": 0.1773720234632492,
|
| 5 |
+
"eval_runtime": 20.504,
|
| 6 |
+
"eval_samples": 312,
|
| 7 |
+
"eval_samples_per_second": 15.217,
|
| 8 |
+
"eval_steps_per_second": 1.902,
|
| 9 |
+
"eval_wer": 0.08439275087909115
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 42.0,
|
| 3 |
+
"total_flos": 8.489873279367732e+19,
|
| 4 |
+
"train_loss": 0.24954385395291484,
|
| 5 |
+
"train_runtime": 14687.2421,
|
| 6 |
+
"train_samples": 2517,
|
| 7 |
+
"train_samples_per_second": 17.137,
|
| 8 |
+
"train_steps_per_second": 0.538
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,649 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.1773720234632492,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-xlsr-1b-mecita-portuguese-all-text-a_coisa-os_morcegos/checkpoint-1738",
|
| 4 |
+
"epoch": 42.0,
|
| 5 |
+
"global_step": 3318,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 3e-05,
|
| 13 |
+
"loss": 25.5905,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_cer": 0.0734213636130093,
|
| 19 |
+
"eval_loss": 0.44948670268058777,
|
| 20 |
+
"eval_runtime": 19.9817,
|
| 21 |
+
"eval_samples_per_second": 15.614,
|
| 22 |
+
"eval_steps_per_second": 1.952,
|
| 23 |
+
"eval_wer": 0.25804706518799025,
|
| 24 |
+
"step": 79
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.27,
|
| 28 |
+
"learning_rate": 2.9631645569620254e-05,
|
| 29 |
+
"loss": 3.1482,
|
| 30 |
+
"step": 100
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 2.0,
|
| 34 |
+
"eval_cer": 0.038020860093510765,
|
| 35 |
+
"eval_loss": 0.2479224056005478,
|
| 36 |
+
"eval_runtime": 19.8794,
|
| 37 |
+
"eval_samples_per_second": 15.695,
|
| 38 |
+
"eval_steps_per_second": 1.962,
|
| 39 |
+
"eval_wer": 0.12036786583716527,
|
| 40 |
+
"step": 158
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 2.53,
|
| 44 |
+
"learning_rate": 2.9251898734177216e-05,
|
| 45 |
+
"loss": 0.4247,
|
| 46 |
+
"step": 200
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 3.0,
|
| 50 |
+
"eval_cer": 0.03447567178749422,
|
| 51 |
+
"eval_loss": 0.23473307490348816,
|
| 52 |
+
"eval_runtime": 19.9054,
|
| 53 |
+
"eval_samples_per_second": 15.674,
|
| 54 |
+
"eval_steps_per_second": 1.959,
|
| 55 |
+
"eval_wer": 0.10251555315120368,
|
| 56 |
+
"step": 237
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 3.8,
|
| 60 |
+
"learning_rate": 2.887215189873418e-05,
|
| 61 |
+
"loss": 0.3136,
|
| 62 |
+
"step": 300
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 4.0,
|
| 66 |
+
"eval_cer": 0.032214971998150335,
|
| 67 |
+
"eval_loss": 0.2043968290090561,
|
| 68 |
+
"eval_runtime": 20.064,
|
| 69 |
+
"eval_samples_per_second": 15.55,
|
| 70 |
+
"eval_steps_per_second": 1.944,
|
| 71 |
+
"eval_wer": 0.10170408439275087,
|
| 72 |
+
"step": 316
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 5.0,
|
| 76 |
+
"eval_cer": 0.029594615424138108,
|
| 77 |
+
"eval_loss": 0.19060839712619781,
|
| 78 |
+
"eval_runtime": 19.9461,
|
| 79 |
+
"eval_samples_per_second": 15.642,
|
| 80 |
+
"eval_steps_per_second": 1.955,
|
| 81 |
+
"eval_wer": 0.09304841763592102,
|
| 82 |
+
"step": 395
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 5.06,
|
| 86 |
+
"learning_rate": 2.849240506329114e-05,
|
| 87 |
+
"loss": 0.2985,
|
| 88 |
+
"step": 400
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 6.0,
|
| 92 |
+
"eval_cer": 0.031084622103478395,
|
| 93 |
+
"eval_loss": 0.2050233781337738,
|
| 94 |
+
"eval_runtime": 20.0744,
|
| 95 |
+
"eval_samples_per_second": 15.542,
|
| 96 |
+
"eval_steps_per_second": 1.943,
|
| 97 |
+
"eval_wer": 0.09629429266973222,
|
| 98 |
+
"step": 474
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.33,
|
| 102 |
+
"learning_rate": 2.8112658227848103e-05,
|
| 103 |
+
"loss": 0.2413,
|
| 104 |
+
"step": 500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 7.0,
|
| 108 |
+
"eval_cer": 0.030879103940810768,
|
| 109 |
+
"eval_loss": 0.20249603688716888,
|
| 110 |
+
"eval_runtime": 20.1379,
|
| 111 |
+
"eval_samples_per_second": 15.493,
|
| 112 |
+
"eval_steps_per_second": 1.937,
|
| 113 |
+
"eval_wer": 0.09710576142818501,
|
| 114 |
+
"step": 553
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 7.59,
|
| 118 |
+
"learning_rate": 2.7732911392405062e-05,
|
| 119 |
+
"loss": 0.2267,
|
| 120 |
+
"step": 600
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 8.0,
|
| 124 |
+
"eval_cer": 0.02913219955813595,
|
| 125 |
+
"eval_loss": 0.20055869221687317,
|
| 126 |
+
"eval_runtime": 19.8715,
|
| 127 |
+
"eval_samples_per_second": 15.701,
|
| 128 |
+
"eval_steps_per_second": 1.963,
|
| 129 |
+
"eval_wer": 0.08845009467135516,
|
| 130 |
+
"step": 632
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 8.86,
|
| 134 |
+
"learning_rate": 2.7353164556962024e-05,
|
| 135 |
+
"loss": 0.224,
|
| 136 |
+
"step": 700
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 9.0,
|
| 140 |
+
"eval_cer": 0.029080820017469044,
|
| 141 |
+
"eval_loss": 0.1990896761417389,
|
| 142 |
+
"eval_runtime": 19.8842,
|
| 143 |
+
"eval_samples_per_second": 15.691,
|
| 144 |
+
"eval_steps_per_second": 1.961,
|
| 145 |
+
"eval_wer": 0.09169596970516636,
|
| 146 |
+
"step": 711
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 10.0,
|
| 150 |
+
"eval_cer": 0.02810460874479782,
|
| 151 |
+
"eval_loss": 0.1880808174610138,
|
| 152 |
+
"eval_runtime": 19.8781,
|
| 153 |
+
"eval_samples_per_second": 15.696,
|
| 154 |
+
"eval_steps_per_second": 1.962,
|
| 155 |
+
"eval_wer": 0.08845009467135516,
|
| 156 |
+
"step": 790
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 10.13,
|
| 160 |
+
"learning_rate": 2.697341772151899e-05,
|
| 161 |
+
"loss": 0.1864,
|
| 162 |
+
"step": 800
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 11.0,
|
| 166 |
+
"eval_cer": 0.02784771104146329,
|
| 167 |
+
"eval_loss": 0.18414445221424103,
|
| 168 |
+
"eval_runtime": 20.0151,
|
| 169 |
+
"eval_samples_per_second": 15.588,
|
| 170 |
+
"eval_steps_per_second": 1.949,
|
| 171 |
+
"eval_wer": 0.08926156342980796,
|
| 172 |
+
"step": 869
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 11.39,
|
| 176 |
+
"learning_rate": 2.659367088607595e-05,
|
| 177 |
+
"loss": 0.1951,
|
| 178 |
+
"step": 900
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 12.0,
|
| 182 |
+
"eval_cer": 0.028155988285464728,
|
| 183 |
+
"eval_loss": 0.18085478246212006,
|
| 184 |
+
"eval_runtime": 20.0956,
|
| 185 |
+
"eval_samples_per_second": 15.526,
|
| 186 |
+
"eval_steps_per_second": 1.941,
|
| 187 |
+
"eval_wer": 0.08953205301595889,
|
| 188 |
+
"step": 948
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 12.66,
|
| 192 |
+
"learning_rate": 2.621392405063291e-05,
|
| 193 |
+
"loss": 0.1794,
|
| 194 |
+
"step": 1000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 13.0,
|
| 198 |
+
"eval_cer": 0.028001849663464008,
|
| 199 |
+
"eval_loss": 0.19231769442558289,
|
| 200 |
+
"eval_runtime": 20.0602,
|
| 201 |
+
"eval_samples_per_second": 15.553,
|
| 202 |
+
"eval_steps_per_second": 1.944,
|
| 203 |
+
"eval_wer": 0.08331079253448742,
|
| 204 |
+
"step": 1027
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 13.92,
|
| 208 |
+
"learning_rate": 2.5834177215189873e-05,
|
| 209 |
+
"loss": 0.1621,
|
| 210 |
+
"step": 1100
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 14.0,
|
| 214 |
+
"eval_cer": 0.027744951960129476,
|
| 215 |
+
"eval_loss": 0.19489158689975739,
|
| 216 |
+
"eval_runtime": 20.0375,
|
| 217 |
+
"eval_samples_per_second": 15.571,
|
| 218 |
+
"eval_steps_per_second": 1.946,
|
| 219 |
+
"eval_wer": 0.08574519880984582,
|
| 220 |
+
"step": 1106
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 15.0,
|
| 224 |
+
"eval_cer": 0.02656322252479063,
|
| 225 |
+
"eval_loss": 0.19287976622581482,
|
| 226 |
+
"eval_runtime": 20.0101,
|
| 227 |
+
"eval_samples_per_second": 15.592,
|
| 228 |
+
"eval_steps_per_second": 1.949,
|
| 229 |
+
"eval_wer": 0.08168785501758183,
|
| 230 |
+
"step": 1185
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 15.19,
|
| 234 |
+
"learning_rate": 2.5454430379746836e-05,
|
| 235 |
+
"loss": 0.1695,
|
| 236 |
+
"step": 1200
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 16.0,
|
| 240 |
+
"eval_cer": 0.02697425885012588,
|
| 241 |
+
"eval_loss": 0.19071535766124725,
|
| 242 |
+
"eval_runtime": 20.1621,
|
| 243 |
+
"eval_samples_per_second": 15.475,
|
| 244 |
+
"eval_steps_per_second": 1.934,
|
| 245 |
+
"eval_wer": 0.08385177170678929,
|
| 246 |
+
"step": 1264
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 16.46,
|
| 250 |
+
"learning_rate": 2.5074683544303798e-05,
|
| 251 |
+
"loss": 0.1528,
|
| 252 |
+
"step": 1300
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 17.0,
|
| 256 |
+
"eval_cer": 0.02856702461079998,
|
| 257 |
+
"eval_loss": 0.18394820392131805,
|
| 258 |
+
"eval_runtime": 19.8761,
|
| 259 |
+
"eval_samples_per_second": 15.697,
|
| 260 |
+
"eval_steps_per_second": 1.962,
|
| 261 |
+
"eval_wer": 0.09061401136056262,
|
| 262 |
+
"step": 1343
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 17.72,
|
| 266 |
+
"learning_rate": 2.469493670886076e-05,
|
| 267 |
+
"loss": 0.1592,
|
| 268 |
+
"step": 1400
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 18.0,
|
| 272 |
+
"eval_cer": 0.02810460874479782,
|
| 273 |
+
"eval_loss": 0.18656496703624725,
|
| 274 |
+
"eval_runtime": 20.0039,
|
| 275 |
+
"eval_samples_per_second": 15.597,
|
| 276 |
+
"eval_steps_per_second": 1.95,
|
| 277 |
+
"eval_wer": 0.09034352177441168,
|
| 278 |
+
"step": 1422
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 18.99,
|
| 282 |
+
"learning_rate": 2.4315189873417722e-05,
|
| 283 |
+
"loss": 0.1519,
|
| 284 |
+
"step": 1500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 19.0,
|
| 288 |
+
"eval_cer": 0.027488054256794944,
|
| 289 |
+
"eval_loss": 0.20305851101875305,
|
| 290 |
+
"eval_runtime": 20.3338,
|
| 291 |
+
"eval_samples_per_second": 15.344,
|
| 292 |
+
"eval_steps_per_second": 1.918,
|
| 293 |
+
"eval_wer": 0.08574519880984582,
|
| 294 |
+
"step": 1501
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 20.0,
|
| 298 |
+
"eval_cer": 0.027796331500796384,
|
| 299 |
+
"eval_loss": 0.19475489854812622,
|
| 300 |
+
"eval_runtime": 19.9135,
|
| 301 |
+
"eval_samples_per_second": 15.668,
|
| 302 |
+
"eval_steps_per_second": 1.958,
|
| 303 |
+
"eval_wer": 0.08601568839599676,
|
| 304 |
+
"step": 1580
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 20.25,
|
| 308 |
+
"learning_rate": 2.3935443037974685e-05,
|
| 309 |
+
"loss": 0.1257,
|
| 310 |
+
"step": 1600
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 21.0,
|
| 314 |
+
"eval_cer": 0.026152186199455377,
|
| 315 |
+
"eval_loss": 0.18495479226112366,
|
| 316 |
+
"eval_runtime": 19.994,
|
| 317 |
+
"eval_samples_per_second": 15.605,
|
| 318 |
+
"eval_steps_per_second": 1.951,
|
| 319 |
+
"eval_wer": 0.08601568839599676,
|
| 320 |
+
"step": 1659
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 21.52,
|
| 324 |
+
"learning_rate": 2.3555696202531647e-05,
|
| 325 |
+
"loss": 0.1288,
|
| 326 |
+
"step": 1700
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 22.0,
|
| 330 |
+
"eval_cer": 0.02656322252479063,
|
| 331 |
+
"eval_loss": 0.1773720234632492,
|
| 332 |
+
"eval_runtime": 19.9911,
|
| 333 |
+
"eval_samples_per_second": 15.607,
|
| 334 |
+
"eval_steps_per_second": 1.951,
|
| 335 |
+
"eval_wer": 0.08439275087909115,
|
| 336 |
+
"step": 1738
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 22.78,
|
| 340 |
+
"learning_rate": 2.317594936708861e-05,
|
| 341 |
+
"loss": 0.115,
|
| 342 |
+
"step": 1800
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 23.0,
|
| 346 |
+
"eval_cer": 0.02651184298412372,
|
| 347 |
+
"eval_loss": 0.19595499336719513,
|
| 348 |
+
"eval_runtime": 19.9245,
|
| 349 |
+
"eval_samples_per_second": 15.659,
|
| 350 |
+
"eval_steps_per_second": 1.957,
|
| 351 |
+
"eval_wer": 0.08439275087909115,
|
| 352 |
+
"step": 1817
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 24.0,
|
| 356 |
+
"eval_cer": 0.025843908955453937,
|
| 357 |
+
"eval_loss": 0.18322429060935974,
|
| 358 |
+
"eval_runtime": 19.9298,
|
| 359 |
+
"eval_samples_per_second": 15.655,
|
| 360 |
+
"eval_steps_per_second": 1.957,
|
| 361 |
+
"eval_wer": 0.08249932377603462,
|
| 362 |
+
"step": 1896
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 24.05,
|
| 366 |
+
"learning_rate": 2.2796202531645568e-05,
|
| 367 |
+
"loss": 0.1223,
|
| 368 |
+
"step": 1900
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 25.0,
|
| 372 |
+
"eval_cer": 0.02610080665878847,
|
| 373 |
+
"eval_loss": 0.19195543229579926,
|
| 374 |
+
"eval_runtime": 20.0272,
|
| 375 |
+
"eval_samples_per_second": 15.579,
|
| 376 |
+
"eval_steps_per_second": 1.947,
|
| 377 |
+
"eval_wer": 0.08276981336218556,
|
| 378 |
+
"step": 1975
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 25.32,
|
| 382 |
+
"learning_rate": 2.2416455696202534e-05,
|
| 383 |
+
"loss": 0.1175,
|
| 384 |
+
"step": 2000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 26.0,
|
| 388 |
+
"eval_cer": 0.026049427118121565,
|
| 389 |
+
"eval_loss": 0.1951374113559723,
|
| 390 |
+
"eval_runtime": 20.0587,
|
| 391 |
+
"eval_samples_per_second": 15.554,
|
| 392 |
+
"eval_steps_per_second": 1.944,
|
| 393 |
+
"eval_wer": 0.08033540708682715,
|
| 394 |
+
"step": 2054
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 26.58,
|
| 398 |
+
"learning_rate": 2.2036708860759496e-05,
|
| 399 |
+
"loss": 0.1051,
|
| 400 |
+
"step": 2100
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 27.0,
|
| 404 |
+
"eval_cer": 0.02656322252479063,
|
| 405 |
+
"eval_loss": 0.1995711475610733,
|
| 406 |
+
"eval_runtime": 20.0348,
|
| 407 |
+
"eval_samples_per_second": 15.573,
|
| 408 |
+
"eval_steps_per_second": 1.947,
|
| 409 |
+
"eval_wer": 0.08249932377603462,
|
| 410 |
+
"step": 2133
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 27.85,
|
| 414 |
+
"learning_rate": 2.1656962025316455e-05,
|
| 415 |
+
"loss": 0.1033,
|
| 416 |
+
"step": 2200
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 28.0,
|
| 420 |
+
"eval_cer": 0.027385295175461132,
|
| 421 |
+
"eval_loss": 0.21518437564373016,
|
| 422 |
+
"eval_runtime": 20.185,
|
| 423 |
+
"eval_samples_per_second": 15.457,
|
| 424 |
+
"eval_steps_per_second": 1.932,
|
| 425 |
+
"eval_wer": 0.08466324046524208,
|
| 426 |
+
"step": 2212
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 29.0,
|
| 430 |
+
"eval_cer": 0.02769357241946257,
|
| 431 |
+
"eval_loss": 0.2082461565732956,
|
| 432 |
+
"eval_runtime": 20.1321,
|
| 433 |
+
"eval_samples_per_second": 15.498,
|
| 434 |
+
"eval_steps_per_second": 1.937,
|
| 435 |
+
"eval_wer": 0.08790911549905328,
|
| 436 |
+
"step": 2291
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 29.11,
|
| 440 |
+
"learning_rate": 2.127721518987342e-05,
|
| 441 |
+
"loss": 0.0961,
|
| 442 |
+
"step": 2300
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 30.0,
|
| 446 |
+
"eval_cer": 0.027385295175461132,
|
| 447 |
+
"eval_loss": 0.2153148353099823,
|
| 448 |
+
"eval_runtime": 20.1045,
|
| 449 |
+
"eval_samples_per_second": 15.519,
|
| 450 |
+
"eval_steps_per_second": 1.94,
|
| 451 |
+
"eval_wer": 0.08547470922369489,
|
| 452 |
+
"step": 2370
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 30.38,
|
| 456 |
+
"learning_rate": 2.089746835443038e-05,
|
| 457 |
+
"loss": 0.1003,
|
| 458 |
+
"step": 2400
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 31.0,
|
| 462 |
+
"eval_cer": 0.028772542773467604,
|
| 463 |
+
"eval_loss": 0.2043757140636444,
|
| 464 |
+
"eval_runtime": 20.1868,
|
| 465 |
+
"eval_samples_per_second": 15.456,
|
| 466 |
+
"eval_steps_per_second": 1.932,
|
| 467 |
+
"eval_wer": 0.09034352177441168,
|
| 468 |
+
"step": 2449
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 31.65,
|
| 472 |
+
"learning_rate": 2.0517721518987342e-05,
|
| 473 |
+
"loss": 0.1129,
|
| 474 |
+
"step": 2500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 32.0,
|
| 478 |
+
"eval_cer": 0.026768740687458253,
|
| 479 |
+
"eval_loss": 0.20504404604434967,
|
| 480 |
+
"eval_runtime": 20.2156,
|
| 481 |
+
"eval_samples_per_second": 15.434,
|
| 482 |
+
"eval_steps_per_second": 1.929,
|
| 483 |
+
"eval_wer": 0.08547470922369489,
|
| 484 |
+
"step": 2528
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 32.91,
|
| 488 |
+
"learning_rate": 2.0137974683544304e-05,
|
| 489 |
+
"loss": 0.0939,
|
| 490 |
+
"step": 2600
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 33.0,
|
| 494 |
+
"eval_cer": 0.0271283974721266,
|
| 495 |
+
"eval_loss": 0.20279404520988464,
|
| 496 |
+
"eval_runtime": 20.1612,
|
| 497 |
+
"eval_samples_per_second": 15.475,
|
| 498 |
+
"eval_steps_per_second": 1.934,
|
| 499 |
+
"eval_wer": 0.08601568839599676,
|
| 500 |
+
"step": 2607
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 34.0,
|
| 504 |
+
"eval_cer": 0.027436674716128037,
|
| 505 |
+
"eval_loss": 0.20313425362110138,
|
| 506 |
+
"eval_runtime": 20.5421,
|
| 507 |
+
"eval_samples_per_second": 15.188,
|
| 508 |
+
"eval_steps_per_second": 1.899,
|
| 509 |
+
"eval_wer": 0.08466324046524208,
|
| 510 |
+
"step": 2686
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 34.18,
|
| 514 |
+
"learning_rate": 1.9758227848101266e-05,
|
| 515 |
+
"loss": 0.0846,
|
| 516 |
+
"step": 2700
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 35.0,
|
| 520 |
+
"eval_cer": 0.026922879309458973,
|
| 521 |
+
"eval_loss": 0.20462077856063843,
|
| 522 |
+
"eval_runtime": 20.1988,
|
| 523 |
+
"eval_samples_per_second": 15.446,
|
| 524 |
+
"eval_steps_per_second": 1.931,
|
| 525 |
+
"eval_wer": 0.08222883418988369,
|
| 526 |
+
"step": 2765
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 35.44,
|
| 530 |
+
"learning_rate": 1.937848101265823e-05,
|
| 531 |
+
"loss": 0.083,
|
| 532 |
+
"step": 2800
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 36.0,
|
| 536 |
+
"eval_cer": 0.02651184298412372,
|
| 537 |
+
"eval_loss": 0.20938025414943695,
|
| 538 |
+
"eval_runtime": 20.4858,
|
| 539 |
+
"eval_samples_per_second": 15.23,
|
| 540 |
+
"eval_steps_per_second": 1.904,
|
| 541 |
+
"eval_wer": 0.08249932377603462,
|
| 542 |
+
"step": 2844
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 36.71,
|
| 546 |
+
"learning_rate": 1.8998734177215188e-05,
|
| 547 |
+
"loss": 0.0844,
|
| 548 |
+
"step": 2900
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 37.0,
|
| 552 |
+
"eval_cer": 0.026768740687458253,
|
| 553 |
+
"eval_loss": 0.21756267547607422,
|
| 554 |
+
"eval_runtime": 20.1565,
|
| 555 |
+
"eval_samples_per_second": 15.479,
|
| 556 |
+
"eval_steps_per_second": 1.935,
|
| 557 |
+
"eval_wer": 0.08195834460373276,
|
| 558 |
+
"step": 2923
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 37.97,
|
| 562 |
+
"learning_rate": 1.8618987341772153e-05,
|
| 563 |
+
"loss": 0.0829,
|
| 564 |
+
"step": 3000
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 38.0,
|
| 568 |
+
"eval_cer": 0.02666598160612444,
|
| 569 |
+
"eval_loss": 0.20818965137004852,
|
| 570 |
+
"eval_runtime": 20.1541,
|
| 571 |
+
"eval_samples_per_second": 15.481,
|
| 572 |
+
"eval_steps_per_second": 1.935,
|
| 573 |
+
"eval_wer": 0.08168785501758183,
|
| 574 |
+
"step": 3002
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 39.0,
|
| 578 |
+
"eval_cer": 0.028618404151466884,
|
| 579 |
+
"eval_loss": 0.2200096845626831,
|
| 580 |
+
"eval_runtime": 20.1838,
|
| 581 |
+
"eval_samples_per_second": 15.458,
|
| 582 |
+
"eval_steps_per_second": 1.932,
|
| 583 |
+
"eval_wer": 0.08926156342980796,
|
| 584 |
+
"step": 3081
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 39.24,
|
| 588 |
+
"learning_rate": 1.8239240506329115e-05,
|
| 589 |
+
"loss": 0.103,
|
| 590 |
+
"step": 3100
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 40.0,
|
| 594 |
+
"eval_cer": 0.027642192878795664,
|
| 595 |
+
"eval_loss": 0.2102428525686264,
|
| 596 |
+
"eval_runtime": 20.0863,
|
| 597 |
+
"eval_samples_per_second": 15.533,
|
| 598 |
+
"eval_steps_per_second": 1.942,
|
| 599 |
+
"eval_wer": 0.08412226129294022,
|
| 600 |
+
"step": 3160
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 40.51,
|
| 604 |
+
"learning_rate": 1.7859493670886074e-05,
|
| 605 |
+
"loss": 0.0728,
|
| 606 |
+
"step": 3200
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 41.0,
|
| 610 |
+
"eval_cer": 0.027077017931459692,
|
| 611 |
+
"eval_loss": 0.2143275886774063,
|
| 612 |
+
"eval_runtime": 20.2051,
|
| 613 |
+
"eval_samples_per_second": 15.442,
|
| 614 |
+
"eval_steps_per_second": 1.93,
|
| 615 |
+
"eval_wer": 0.08168785501758183,
|
| 616 |
+
"step": 3239
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 41.77,
|
| 620 |
+
"learning_rate": 1.747974683544304e-05,
|
| 621 |
+
"loss": 0.079,
|
| 622 |
+
"step": 3300
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 42.0,
|
| 626 |
+
"eval_cer": 0.02651184298412372,
|
| 627 |
+
"eval_loss": 0.2131078988313675,
|
| 628 |
+
"eval_runtime": 20.1578,
|
| 629 |
+
"eval_samples_per_second": 15.478,
|
| 630 |
+
"eval_steps_per_second": 1.935,
|
| 631 |
+
"eval_wer": 0.08249932377603462,
|
| 632 |
+
"step": 3318
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 42.0,
|
| 636 |
+
"step": 3318,
|
| 637 |
+
"total_flos": 8.489873279367732e+19,
|
| 638 |
+
"train_loss": 0.24954385395291484,
|
| 639 |
+
"train_runtime": 14687.2421,
|
| 640 |
+
"train_samples_per_second": 17.137,
|
| 641 |
+
"train_steps_per_second": 0.538
|
| 642 |
+
}
|
| 643 |
+
],
|
| 644 |
+
"max_steps": 7900,
|
| 645 |
+
"num_train_epochs": 100,
|
| 646 |
+
"total_flos": 8.489873279367732e+19,
|
| 647 |
+
"trial_name": null,
|
| 648 |
+
"trial_params": null
|
| 649 |
+
}
|