Commit
•
334d55c
1
Parent(s):
ff76b19
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 41.67,
|
| 3 |
+
"eval_loss": 0.74560546875,
|
| 4 |
+
"eval_runtime": 18.2949,
|
| 5 |
+
"eval_samples_per_second": 3.061,
|
| 6 |
+
"eval_steps_per_second": 0.109,
|
| 7 |
+
"eval_wer": 18.448637316561843,
|
| 8 |
+
"train_loss": 0.05835451430082321,
|
| 9 |
+
"train_runtime": 9110.0805,
|
| 10 |
+
"train_samples_per_second": 7.025,
|
| 11 |
+
"train_steps_per_second": 0.055
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 41.67,
|
| 3 |
+
"eval_loss": 0.74560546875,
|
| 4 |
+
"eval_runtime": 18.2949,
|
| 5 |
+
"eval_samples_per_second": 3.061,
|
| 6 |
+
"eval_steps_per_second": 0.109,
|
| 7 |
+
"eval_wer": 18.448637316561843
|
| 8 |
+
}
|
runs/Dec22_09-29-33_bm-whisper-a10/events.out.tfevents.1671710931.bm-whisper-a10.24020.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e52e5baea847e1ef4199e7eb070a827f22b0217d2d7c3412bfb561950d115cb
|
| 3 |
+
size 358
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 41.67,
|
| 3 |
+
"train_loss": 0.05835451430082321,
|
| 4 |
+
"train_runtime": 9110.0805,
|
| 5 |
+
"train_samples_per_second": 7.025,
|
| 6 |
+
"train_steps_per_second": 0.055
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,190 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 18.448637316561843,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-200",
|
| 4 |
+
"epoch": 41.666666666666664,
|
| 5 |
+
"global_step": 500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 2.08,
|
| 12 |
+
"learning_rate": 5.746206252093564e-06,
|
| 13 |
+
"loss": 1.0017,
|
| 14 |
+
"step": 25
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 4.17,
|
| 18 |
+
"learning_rate": 7.226145833886759e-06,
|
| 19 |
+
"loss": 0.1431,
|
| 20 |
+
"step": 50
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 6.25,
|
| 24 |
+
"learning_rate": 8.045346441409077e-06,
|
| 25 |
+
"loss": 0.0125,
|
| 26 |
+
"step": 75
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 8.33,
|
| 30 |
+
"learning_rate": 8.614712701594175e-06,
|
| 31 |
+
"loss": 0.0029,
|
| 32 |
+
"step": 100
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 8.33,
|
| 36 |
+
"eval_loss": 0.6650390625,
|
| 37 |
+
"eval_runtime": 18.7793,
|
| 38 |
+
"eval_samples_per_second": 2.982,
|
| 39 |
+
"eval_steps_per_second": 0.107,
|
| 40 |
+
"eval_wer": 19.28721174004193,
|
| 41 |
+
"step": 100
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 10.42,
|
| 45 |
+
"learning_rate": 9.051535070126044e-06,
|
| 46 |
+
"loss": 0.0014,
|
| 47 |
+
"step": 125
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 12.5,
|
| 51 |
+
"learning_rate": 9.406017565450707e-06,
|
| 52 |
+
"loss": 0.0009,
|
| 53 |
+
"step": 150
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 14.58,
|
| 57 |
+
"learning_rate": 9.704332905698628e-06,
|
| 58 |
+
"loss": 0.0007,
|
| 59 |
+
"step": 175
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 16.67,
|
| 63 |
+
"learning_rate": 9.961869578736312e-06,
|
| 64 |
+
"loss": 0.0005,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 16.67,
|
| 69 |
+
"eval_loss": 0.74560546875,
|
| 70 |
+
"eval_runtime": 19.1155,
|
| 71 |
+
"eval_samples_per_second": 2.93,
|
| 72 |
+
"eval_steps_per_second": 0.105,
|
| 73 |
+
"eval_wer": 18.448637316561843,
|
| 74 |
+
"step": 200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 18.75,
|
| 78 |
+
"learning_rate": 9.333333333333334e-06,
|
| 79 |
+
"loss": 0.0004,
|
| 80 |
+
"step": 225
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 20.83,
|
| 84 |
+
"learning_rate": 8.5e-06,
|
| 85 |
+
"loss": 0.0004,
|
| 86 |
+
"step": 250
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 22.92,
|
| 90 |
+
"learning_rate": 7.666666666666667e-06,
|
| 91 |
+
"loss": 0.0003,
|
| 92 |
+
"step": 275
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 25.0,
|
| 96 |
+
"learning_rate": 6.833333333333334e-06,
|
| 97 |
+
"loss": 0.0003,
|
| 98 |
+
"step": 300
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 25.0,
|
| 102 |
+
"eval_loss": 0.77978515625,
|
| 103 |
+
"eval_runtime": 19.0324,
|
| 104 |
+
"eval_samples_per_second": 2.942,
|
| 105 |
+
"eval_steps_per_second": 0.105,
|
| 106 |
+
"eval_wer": 19.49685534591195,
|
| 107 |
+
"step": 300
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 27.08,
|
| 111 |
+
"learning_rate": 6e-06,
|
| 112 |
+
"loss": 0.0003,
|
| 113 |
+
"step": 325
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 29.17,
|
| 117 |
+
"learning_rate": 5.1666666666666675e-06,
|
| 118 |
+
"loss": 0.0003,
|
| 119 |
+
"step": 350
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 31.25,
|
| 123 |
+
"learning_rate": 4.333333333333334e-06,
|
| 124 |
+
"loss": 0.0002,
|
| 125 |
+
"step": 375
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 33.33,
|
| 129 |
+
"learning_rate": 3.5e-06,
|
| 130 |
+
"loss": 0.0002,
|
| 131 |
+
"step": 400
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 33.33,
|
| 135 |
+
"eval_loss": 0.79638671875,
|
| 136 |
+
"eval_runtime": 18.9232,
|
| 137 |
+
"eval_samples_per_second": 2.959,
|
| 138 |
+
"eval_steps_per_second": 0.106,
|
| 139 |
+
"eval_wer": 19.70649895178197,
|
| 140 |
+
"step": 400
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 35.42,
|
| 144 |
+
"learning_rate": 2.666666666666667e-06,
|
| 145 |
+
"loss": 0.0002,
|
| 146 |
+
"step": 425
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 37.5,
|
| 150 |
+
"learning_rate": 1.8333333333333333e-06,
|
| 151 |
+
"loss": 0.0002,
|
| 152 |
+
"step": 450
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 39.58,
|
| 156 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 157 |
+
"loss": 0.0002,
|
| 158 |
+
"step": 475
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 41.67,
|
| 162 |
+
"learning_rate": 1.6666666666666668e-07,
|
| 163 |
+
"loss": 0.0002,
|
| 164 |
+
"step": 500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 41.67,
|
| 168 |
+
"eval_loss": 0.80224609375,
|
| 169 |
+
"eval_runtime": 19.0419,
|
| 170 |
+
"eval_samples_per_second": 2.941,
|
| 171 |
+
"eval_steps_per_second": 0.105,
|
| 172 |
+
"eval_wer": 20.020964360587,
|
| 173 |
+
"step": 500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 41.67,
|
| 177 |
+
"step": 500,
|
| 178 |
+
"total_flos": 1.3379374861974818e+20,
|
| 179 |
+
"train_loss": 0.05835451430082321,
|
| 180 |
+
"train_runtime": 9110.0805,
|
| 181 |
+
"train_samples_per_second": 7.025,
|
| 182 |
+
"train_steps_per_second": 0.055
|
| 183 |
+
}
|
| 184 |
+
],
|
| 185 |
+
"max_steps": 500,
|
| 186 |
+
"num_train_epochs": 42,
|
| 187 |
+
"total_flos": 1.3379374861974818e+20,
|
| 188 |
+
"trial_name": null,
|
| 189 |
+
"trial_params": null
|
| 190 |
+
}
|