Commit
·
9734651
1
Parent(s):
bb3e0b4
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- pytorch_model.bin +1 -1
- train_results.json +8 -0
- trainer_state.json +478 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.49949410557746887,
|
| 4 |
+
"eval_runtime": 49.2354,
|
| 5 |
+
"eval_samples": 800,
|
| 6 |
+
"eval_samples_per_second": 16.248,
|
| 7 |
+
"eval_steps_per_second": 16.248,
|
| 8 |
+
"eval_wer": 0.4421082540133542,
|
| 9 |
+
"train_loss": 1.1293389116707495,
|
| 10 |
+
"train_runtime": 13204.1924,
|
| 11 |
+
"train_samples": 1857,
|
| 12 |
+
"train_samples_per_second": 14.064,
|
| 13 |
+
"train_steps_per_second": 0.447
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.49949410557746887,
|
| 4 |
+
"eval_runtime": 49.2354,
|
| 5 |
+
"eval_samples": 800,
|
| 6 |
+
"eval_samples_per_second": 16.248,
|
| 7 |
+
"eval_steps_per_second": 16.248,
|
| 8 |
+
"eval_wer": 0.4421082540133542
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262124593
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:402656f3d141939c3e93c67f497e74413d401b0038026793dd53d60129337c5d
|
| 3 |
size 1262124593
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 1.1293389116707495,
|
| 4 |
+
"train_runtime": 13204.1924,
|
| 5 |
+
"train_samples": 1857,
|
| 6 |
+
"train_samples_per_second": 14.064,
|
| 7 |
+
"train_steps_per_second": 0.447
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,478 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"global_step": 5900,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.69,
|
| 12 |
+
"learning_rate": 5.88e-05,
|
| 13 |
+
"loss": 7.6789,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 3.39,
|
| 18 |
+
"learning_rate": 0.0001188,
|
| 19 |
+
"loss": 3.2952,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 5.08,
|
| 24 |
+
"learning_rate": 0.00017879999999999998,
|
| 25 |
+
"loss": 3.0495,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 6.78,
|
| 30 |
+
"learning_rate": 0.0002388,
|
| 31 |
+
"loss": 2.7948,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 8.47,
|
| 36 |
+
"learning_rate": 0.0002988,
|
| 37 |
+
"loss": 1.8597,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 8.47,
|
| 42 |
+
"eval_loss": 0.7730758786201477,
|
| 43 |
+
"eval_runtime": 49.5825,
|
| 44 |
+
"eval_samples_per_second": 16.135,
|
| 45 |
+
"eval_steps_per_second": 16.135,
|
| 46 |
+
"eval_wer": 0.7211251598238386,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 10.17,
|
| 51 |
+
"learning_rate": 0.00029455555555555555,
|
| 52 |
+
"loss": 1.4933,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 11.86,
|
| 57 |
+
"learning_rate": 0.000289,
|
| 58 |
+
"loss": 1.365,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 13.56,
|
| 63 |
+
"learning_rate": 0.0002834444444444444,
|
| 64 |
+
"loss": 1.3187,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 15.25,
|
| 69 |
+
"learning_rate": 0.0002778888888888889,
|
| 70 |
+
"loss": 1.2744,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 16.95,
|
| 75 |
+
"learning_rate": 0.0002723333333333333,
|
| 76 |
+
"loss": 1.2508,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 16.95,
|
| 81 |
+
"eval_loss": 0.5367600321769714,
|
| 82 |
+
"eval_runtime": 48.7576,
|
| 83 |
+
"eval_samples_per_second": 16.408,
|
| 84 |
+
"eval_steps_per_second": 16.408,
|
| 85 |
+
"eval_wer": 0.5989487143060094,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 18.64,
|
| 90 |
+
"learning_rate": 0.0002667777777777778,
|
| 91 |
+
"loss": 1.1856,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 20.34,
|
| 96 |
+
"learning_rate": 0.0002612222222222222,
|
| 97 |
+
"loss": 1.1821,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 22.03,
|
| 102 |
+
"learning_rate": 0.00025566666666666663,
|
| 103 |
+
"loss": 1.1284,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 23.73,
|
| 108 |
+
"learning_rate": 0.0002501111111111111,
|
| 109 |
+
"loss": 1.1204,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 25.42,
|
| 114 |
+
"learning_rate": 0.00024455555555555553,
|
| 115 |
+
"loss": 1.1066,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 25.42,
|
| 120 |
+
"eval_loss": 0.5033634901046753,
|
| 121 |
+
"eval_runtime": 48.524,
|
| 122 |
+
"eval_samples_per_second": 16.487,
|
| 123 |
+
"eval_steps_per_second": 16.487,
|
| 124 |
+
"eval_wer": 0.553345645688308,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 27.12,
|
| 129 |
+
"learning_rate": 0.00023905555555555553,
|
| 130 |
+
"loss": 1.0826,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 28.81,
|
| 135 |
+
"learning_rate": 0.00023349999999999998,
|
| 136 |
+
"loss": 1.0375,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 30.51,
|
| 141 |
+
"learning_rate": 0.00022794444444444443,
|
| 142 |
+
"loss": 1.0304,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 32.2,
|
| 147 |
+
"learning_rate": 0.00022238888888888889,
|
| 148 |
+
"loss": 1.0207,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 33.9,
|
| 153 |
+
"learning_rate": 0.0002168333333333333,
|
| 154 |
+
"loss": 1.0064,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 33.9,
|
| 159 |
+
"eval_loss": 0.46856507658958435,
|
| 160 |
+
"eval_runtime": 48.9718,
|
| 161 |
+
"eval_samples_per_second": 16.336,
|
| 162 |
+
"eval_steps_per_second": 16.336,
|
| 163 |
+
"eval_wer": 0.5114362835630061,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 35.59,
|
| 168 |
+
"learning_rate": 0.00021127777777777776,
|
| 169 |
+
"loss": 0.9979,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 37.29,
|
| 174 |
+
"learning_rate": 0.0002057222222222222,
|
| 175 |
+
"loss": 0.9902,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 38.98,
|
| 180 |
+
"learning_rate": 0.00020016666666666666,
|
| 181 |
+
"loss": 0.9511,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 40.68,
|
| 186 |
+
"learning_rate": 0.00019461111111111109,
|
| 187 |
+
"loss": 0.9584,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 42.37,
|
| 192 |
+
"learning_rate": 0.00018905555555555554,
|
| 193 |
+
"loss": 0.9324,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 42.37,
|
| 198 |
+
"eval_loss": 0.49271178245544434,
|
| 199 |
+
"eval_runtime": 48.7262,
|
| 200 |
+
"eval_samples_per_second": 16.418,
|
| 201 |
+
"eval_steps_per_second": 16.418,
|
| 202 |
+
"eval_wer": 0.5056115925557607,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 44.07,
|
| 207 |
+
"learning_rate": 0.0001835,
|
| 208 |
+
"loss": 0.9287,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 45.76,
|
| 213 |
+
"learning_rate": 0.00017794444444444444,
|
| 214 |
+
"loss": 0.9264,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 47.46,
|
| 219 |
+
"learning_rate": 0.00017238888888888886,
|
| 220 |
+
"loss": 0.9031,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 49.15,
|
| 225 |
+
"learning_rate": 0.00016683333333333331,
|
| 226 |
+
"loss": 0.8996,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 50.85,
|
| 231 |
+
"learning_rate": 0.00016127777777777776,
|
| 232 |
+
"loss": 0.876,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 50.85,
|
| 237 |
+
"eval_loss": 0.4733906686306,
|
| 238 |
+
"eval_runtime": 48.5334,
|
| 239 |
+
"eval_samples_per_second": 16.484,
|
| 240 |
+
"eval_steps_per_second": 16.484,
|
| 241 |
+
"eval_wer": 0.4794715158403182,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 52.54,
|
| 246 |
+
"learning_rate": 0.0001557222222222222,
|
| 247 |
+
"loss": 0.8432,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 54.24,
|
| 252 |
+
"learning_rate": 0.00015016666666666664,
|
| 253 |
+
"loss": 0.8592,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 55.93,
|
| 258 |
+
"learning_rate": 0.0001446111111111111,
|
| 259 |
+
"loss": 0.8322,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 57.63,
|
| 264 |
+
"learning_rate": 0.00013905555555555554,
|
| 265 |
+
"loss": 0.8286,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 59.32,
|
| 270 |
+
"learning_rate": 0.0001335,
|
| 271 |
+
"loss": 0.8082,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 59.32,
|
| 276 |
+
"eval_loss": 0.474797785282135,
|
| 277 |
+
"eval_runtime": 48.7764,
|
| 278 |
+
"eval_samples_per_second": 16.401,
|
| 279 |
+
"eval_steps_per_second": 16.401,
|
| 280 |
+
"eval_wer": 0.4798977127432874,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 61.02,
|
| 285 |
+
"learning_rate": 0.00012794444444444442,
|
| 286 |
+
"loss": 0.8101,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 62.71,
|
| 291 |
+
"learning_rate": 0.0001223888888888889,
|
| 292 |
+
"loss": 0.7805,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 64.41,
|
| 297 |
+
"learning_rate": 0.00011683333333333332,
|
| 298 |
+
"loss": 0.7741,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 66.1,
|
| 303 |
+
"learning_rate": 0.00011127777777777777,
|
| 304 |
+
"loss": 0.773,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 67.8,
|
| 309 |
+
"learning_rate": 0.0001057222222222222,
|
| 310 |
+
"loss": 0.7604,
|
| 311 |
+
"step": 4000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 67.8,
|
| 315 |
+
"eval_loss": 0.49494636058807373,
|
| 316 |
+
"eval_runtime": 48.9958,
|
| 317 |
+
"eval_samples_per_second": 16.328,
|
| 318 |
+
"eval_steps_per_second": 16.328,
|
| 319 |
+
"eval_wer": 0.46910072453473506,
|
| 320 |
+
"step": 4000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 69.49,
|
| 324 |
+
"learning_rate": 0.00010016666666666666,
|
| 325 |
+
"loss": 0.7527,
|
| 326 |
+
"step": 4100
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 71.19,
|
| 330 |
+
"learning_rate": 9.46111111111111e-05,
|
| 331 |
+
"loss": 0.7354,
|
| 332 |
+
"step": 4200
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 72.88,
|
| 336 |
+
"learning_rate": 8.905555555555555e-05,
|
| 337 |
+
"loss": 0.7334,
|
| 338 |
+
"step": 4300
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 74.58,
|
| 342 |
+
"learning_rate": 8.349999999999998e-05,
|
| 343 |
+
"loss": 0.7286,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 76.27,
|
| 348 |
+
"learning_rate": 7.794444444444445e-05,
|
| 349 |
+
"loss": 0.7241,
|
| 350 |
+
"step": 4500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 76.27,
|
| 354 |
+
"eval_loss": 0.5090368390083313,
|
| 355 |
+
"eval_runtime": 49.0851,
|
| 356 |
+
"eval_samples_per_second": 16.298,
|
| 357 |
+
"eval_steps_per_second": 16.298,
|
| 358 |
+
"eval_wer": 0.4627077709901975,
|
| 359 |
+
"step": 4500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 77.97,
|
| 363 |
+
"learning_rate": 7.238888888888889e-05,
|
| 364 |
+
"loss": 0.7065,
|
| 365 |
+
"step": 4600
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 79.66,
|
| 369 |
+
"learning_rate": 6.683333333333332e-05,
|
| 370 |
+
"loss": 0.6978,
|
| 371 |
+
"step": 4700
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 81.36,
|
| 375 |
+
"learning_rate": 6.127777777777777e-05,
|
| 376 |
+
"loss": 0.688,
|
| 377 |
+
"step": 4800
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 83.05,
|
| 381 |
+
"learning_rate": 5.572222222222222e-05,
|
| 382 |
+
"loss": 0.6725,
|
| 383 |
+
"step": 4900
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 84.75,
|
| 387 |
+
"learning_rate": 5.016666666666666e-05,
|
| 388 |
+
"loss": 0.6739,
|
| 389 |
+
"step": 5000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 84.75,
|
| 393 |
+
"eval_loss": 0.49671733379364014,
|
| 394 |
+
"eval_runtime": 50.1856,
|
| 395 |
+
"eval_samples_per_second": 15.941,
|
| 396 |
+
"eval_steps_per_second": 15.941,
|
| 397 |
+
"eval_wer": 0.4452336979684614,
|
| 398 |
+
"step": 5000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 86.44,
|
| 402 |
+
"learning_rate": 4.4611111111111106e-05,
|
| 403 |
+
"loss": 0.6675,
|
| 404 |
+
"step": 5100
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 88.14,
|
| 408 |
+
"learning_rate": 3.905555555555555e-05,
|
| 409 |
+
"loss": 0.6548,
|
| 410 |
+
"step": 5200
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 89.83,
|
| 414 |
+
"learning_rate": 3.3499999999999994e-05,
|
| 415 |
+
"loss": 0.6465,
|
| 416 |
+
"step": 5300
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 91.53,
|
| 420 |
+
"learning_rate": 2.8e-05,
|
| 421 |
+
"loss": 0.6471,
|
| 422 |
+
"step": 5400
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 93.22,
|
| 426 |
+
"learning_rate": 2.2444444444444444e-05,
|
| 427 |
+
"loss": 0.6447,
|
| 428 |
+
"step": 5500
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 93.22,
|
| 432 |
+
"eval_loss": 0.5071204900741577,
|
| 433 |
+
"eval_runtime": 50.5031,
|
| 434 |
+
"eval_samples_per_second": 15.841,
|
| 435 |
+
"eval_steps_per_second": 15.841,
|
| 436 |
+
"eval_wer": 0.4436709759909078,
|
| 437 |
+
"step": 5500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 94.92,
|
| 441 |
+
"learning_rate": 1.6888888888888888e-05,
|
| 442 |
+
"loss": 0.6362,
|
| 443 |
+
"step": 5600
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 96.61,
|
| 447 |
+
"learning_rate": 1.1333333333333332e-05,
|
| 448 |
+
"loss": 0.6434,
|
| 449 |
+
"step": 5700
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 98.31,
|
| 453 |
+
"learning_rate": 5.777777777777778e-06,
|
| 454 |
+
"loss": 0.6269,
|
| 455 |
+
"step": 5800
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 100.0,
|
| 459 |
+
"learning_rate": 2.222222222222222e-07,
|
| 460 |
+
"loss": 0.6369,
|
| 461 |
+
"step": 5900
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 100.0,
|
| 465 |
+
"step": 5900,
|
| 466 |
+
"total_flos": 3.4027105250131104e+19,
|
| 467 |
+
"train_loss": 1.1293389116707495,
|
| 468 |
+
"train_runtime": 13204.1924,
|
| 469 |
+
"train_samples_per_second": 14.064,
|
| 470 |
+
"train_steps_per_second": 0.447
|
| 471 |
+
}
|
| 472 |
+
],
|
| 473 |
+
"max_steps": 5900,
|
| 474 |
+
"num_train_epochs": 100,
|
| 475 |
+
"total_flos": 3.4027105250131104e+19,
|
| 476 |
+
"trial_name": null,
|
| 477 |
+
"trial_params": null
|
| 478 |
+
}
|