Commit
•
b11a2b9
1
Parent(s):
89cc42e
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +385 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 18.0,
|
| 3 |
+
"eval_loss": 0.3360544443130493,
|
| 4 |
+
"eval_runtime": 56.6507,
|
| 5 |
+
"eval_samples": 512,
|
| 6 |
+
"eval_samples_per_second": 9.038,
|
| 7 |
+
"eval_steps_per_second": 1.13,
|
| 8 |
+
"eval_wer": 0.19631901840490798,
|
| 9 |
+
"train_loss": 0.4896983008235647,
|
| 10 |
+
"train_runtime": 62213.9785,
|
| 11 |
+
"train_samples": 8619,
|
| 12 |
+
"train_samples_per_second": 2.494,
|
| 13 |
+
"train_steps_per_second": 0.104
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 18.0,
|
| 3 |
+
"eval_loss": 0.3360544443130493,
|
| 4 |
+
"eval_runtime": 56.6507,
|
| 5 |
+
"eval_samples": 512,
|
| 6 |
+
"eval_samples_per_second": 9.038,
|
| 7 |
+
"eval_steps_per_second": 1.13,
|
| 8 |
+
"eval_wer": 0.19631901840490798
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 18.0,
|
| 3 |
+
"train_loss": 0.4896983008235647,
|
| 4 |
+
"train_runtime": 62213.9785,
|
| 5 |
+
"train_samples": 8619,
|
| 6 |
+
"train_samples_per_second": 2.494,
|
| 7 |
+
"train_steps_per_second": 0.104
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,385 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 17.999072356215212,
|
| 5 |
+
"global_step": 6462,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.56,
|
| 12 |
+
"eval_loss": 2.8832294940948486,
|
| 13 |
+
"eval_runtime": 56.6037,
|
| 14 |
+
"eval_samples_per_second": 9.045,
|
| 15 |
+
"eval_steps_per_second": 1.131,
|
| 16 |
+
"eval_wer": 1.0,
|
| 17 |
+
"step": 200
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.11,
|
| 21 |
+
"eval_loss": 1.1704888343811035,
|
| 22 |
+
"eval_runtime": 56.4819,
|
| 23 |
+
"eval_samples_per_second": 9.065,
|
| 24 |
+
"eval_steps_per_second": 1.133,
|
| 25 |
+
"eval_wer": 0.7788489628980426,
|
| 26 |
+
"step": 400
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.39,
|
| 30 |
+
"learning_rate": 0.00029939999999999996,
|
| 31 |
+
"loss": 3.3987,
|
| 32 |
+
"step": 500
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.67,
|
| 36 |
+
"eval_loss": 0.7739295959472656,
|
| 37 |
+
"eval_runtime": 57.0638,
|
| 38 |
+
"eval_samples_per_second": 8.972,
|
| 39 |
+
"eval_steps_per_second": 1.122,
|
| 40 |
+
"eval_wer": 0.5895413380075957,
|
| 41 |
+
"step": 600
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.23,
|
| 45 |
+
"eval_loss": 0.6045235395431519,
|
| 46 |
+
"eval_runtime": 56.2609,
|
| 47 |
+
"eval_samples_per_second": 9.1,
|
| 48 |
+
"eval_steps_per_second": 1.138,
|
| 49 |
+
"eval_wer": 0.4902132632193982,
|
| 50 |
+
"step": 800
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.78,
|
| 54 |
+
"learning_rate": 0.0002748909761824891,
|
| 55 |
+
"loss": 0.8313,
|
| 56 |
+
"step": 1000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.78,
|
| 60 |
+
"eval_loss": 0.5234911441802979,
|
| 61 |
+
"eval_runtime": 56.3102,
|
| 62 |
+
"eval_samples_per_second": 9.092,
|
| 63 |
+
"eval_steps_per_second": 1.137,
|
| 64 |
+
"eval_wer": 0.43938066023955596,
|
| 65 |
+
"step": 1000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 3.34,
|
| 69 |
+
"eval_loss": 0.48238545656204224,
|
| 70 |
+
"eval_runtime": 56.3832,
|
| 71 |
+
"eval_samples_per_second": 9.081,
|
| 72 |
+
"eval_steps_per_second": 1.135,
|
| 73 |
+
"eval_wer": 0.4002337131171487,
|
| 74 |
+
"step": 1200
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 3.9,
|
| 78 |
+
"eval_loss": 0.4378102719783783,
|
| 79 |
+
"eval_runtime": 56.2821,
|
| 80 |
+
"eval_samples_per_second": 9.097,
|
| 81 |
+
"eval_steps_per_second": 1.137,
|
| 82 |
+
"eval_wer": 0.37540169442009935,
|
| 83 |
+
"step": 1400
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 4.18,
|
| 87 |
+
"learning_rate": 0.00024973163367997313,
|
| 88 |
+
"loss": 0.5342,
|
| 89 |
+
"step": 1500
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 4.46,
|
| 93 |
+
"eval_loss": 0.44333964586257935,
|
| 94 |
+
"eval_runtime": 56.3381,
|
| 95 |
+
"eval_samples_per_second": 9.088,
|
| 96 |
+
"eval_steps_per_second": 1.136,
|
| 97 |
+
"eval_wer": 0.36342389716622847,
|
| 98 |
+
"step": 1600
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 5.01,
|
| 102 |
+
"eval_loss": 0.41031020879745483,
|
| 103 |
+
"eval_runtime": 56.4477,
|
| 104 |
+
"eval_samples_per_second": 9.07,
|
| 105 |
+
"eval_steps_per_second": 1.134,
|
| 106 |
+
"eval_wer": 0.34852468594799885,
|
| 107 |
+
"step": 1800
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 5.57,
|
| 111 |
+
"learning_rate": 0.00022457229117745723,
|
| 112 |
+
"loss": 0.3792,
|
| 113 |
+
"step": 2000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 5.57,
|
| 117 |
+
"eval_loss": 0.3816453218460083,
|
| 118 |
+
"eval_runtime": 56.4551,
|
| 119 |
+
"eval_samples_per_second": 9.069,
|
| 120 |
+
"eval_steps_per_second": 1.134,
|
| 121 |
+
"eval_wer": 0.33099620216184633,
|
| 122 |
+
"step": 2000
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 6.13,
|
| 126 |
+
"eval_loss": 0.3952919840812683,
|
| 127 |
+
"eval_runtime": 56.4457,
|
| 128 |
+
"eval_samples_per_second": 9.071,
|
| 129 |
+
"eval_steps_per_second": 1.134,
|
| 130 |
+
"eval_wer": 0.32252410166520595,
|
| 131 |
+
"step": 2200
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 6.68,
|
| 135 |
+
"eval_loss": 0.39945441484451294,
|
| 136 |
+
"eval_runtime": 56.6634,
|
| 137 |
+
"eval_samples_per_second": 9.036,
|
| 138 |
+
"eval_steps_per_second": 1.129,
|
| 139 |
+
"eval_wer": 0.31317557697925796,
|
| 140 |
+
"step": 2400
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 6.96,
|
| 144 |
+
"learning_rate": 0.00019941294867494128,
|
| 145 |
+
"loss": 0.2924,
|
| 146 |
+
"step": 2500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 7.24,
|
| 150 |
+
"eval_loss": 0.3906857967376709,
|
| 151 |
+
"eval_runtime": 56.3831,
|
| 152 |
+
"eval_samples_per_second": 9.081,
|
| 153 |
+
"eval_steps_per_second": 1.135,
|
| 154 |
+
"eval_wer": 0.2930178206251826,
|
| 155 |
+
"step": 2600
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 7.8,
|
| 159 |
+
"eval_loss": 0.35171157121658325,
|
| 160 |
+
"eval_runtime": 56.7082,
|
| 161 |
+
"eval_samples_per_second": 9.029,
|
| 162 |
+
"eval_steps_per_second": 1.129,
|
| 163 |
+
"eval_wer": 0.2740286298568507,
|
| 164 |
+
"step": 2800
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 8.36,
|
| 168 |
+
"learning_rate": 0.00017425360617242535,
|
| 169 |
+
"loss": 0.2217,
|
| 170 |
+
"step": 3000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 8.36,
|
| 174 |
+
"eval_loss": 0.33607447147369385,
|
| 175 |
+
"eval_runtime": 56.5616,
|
| 176 |
+
"eval_samples_per_second": 9.052,
|
| 177 |
+
"eval_steps_per_second": 1.132,
|
| 178 |
+
"eval_wer": 0.2591294186386211,
|
| 179 |
+
"step": 3000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 8.91,
|
| 183 |
+
"eval_loss": 0.3340049088001251,
|
| 184 |
+
"eval_runtime": 56.5741,
|
| 185 |
+
"eval_samples_per_second": 9.05,
|
| 186 |
+
"eval_steps_per_second": 1.131,
|
| 187 |
+
"eval_wer": 0.2451066316096991,
|
| 188 |
+
"step": 3200
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 9.47,
|
| 192 |
+
"eval_loss": 0.3125685453414917,
|
| 193 |
+
"eval_runtime": 56.4726,
|
| 194 |
+
"eval_samples_per_second": 9.066,
|
| 195 |
+
"eval_steps_per_second": 1.133,
|
| 196 |
+
"eval_wer": 0.2448144902132632,
|
| 197 |
+
"step": 3400
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 9.75,
|
| 201 |
+
"learning_rate": 0.00014909426366990943,
|
| 202 |
+
"loss": 0.1714,
|
| 203 |
+
"step": 3500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 10.03,
|
| 207 |
+
"eval_loss": 0.34412676095962524,
|
| 208 |
+
"eval_runtime": 56.5983,
|
| 209 |
+
"eval_samples_per_second": 9.046,
|
| 210 |
+
"eval_steps_per_second": 1.131,
|
| 211 |
+
"eval_wer": 0.2556237218813906,
|
| 212 |
+
"step": 3600
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 10.58,
|
| 216 |
+
"eval_loss": 0.3404456079006195,
|
| 217 |
+
"eval_runtime": 56.5608,
|
| 218 |
+
"eval_samples_per_second": 9.052,
|
| 219 |
+
"eval_steps_per_second": 1.132,
|
| 220 |
+
"eval_wer": 0.2521180251241601,
|
| 221 |
+
"step": 3800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 11.14,
|
| 225 |
+
"learning_rate": 0.0001239349211673935,
|
| 226 |
+
"loss": 0.1395,
|
| 227 |
+
"step": 4000
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 11.14,
|
| 231 |
+
"eval_loss": 0.3728441894054413,
|
| 232 |
+
"eval_runtime": 56.7823,
|
| 233 |
+
"eval_samples_per_second": 9.017,
|
| 234 |
+
"eval_steps_per_second": 1.127,
|
| 235 |
+
"eval_wer": 0.25182588372772424,
|
| 236 |
+
"step": 4000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 11.7,
|
| 240 |
+
"eval_loss": 0.3828706741333008,
|
| 241 |
+
"eval_runtime": 56.4473,
|
| 242 |
+
"eval_samples_per_second": 9.07,
|
| 243 |
+
"eval_steps_per_second": 1.134,
|
| 244 |
+
"eval_wer": 0.23955594507741748,
|
| 245 |
+
"step": 4200
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 12.26,
|
| 249 |
+
"eval_loss": 0.3465881943702698,
|
| 250 |
+
"eval_runtime": 56.3386,
|
| 251 |
+
"eval_samples_per_second": 9.088,
|
| 252 |
+
"eval_steps_per_second": 1.136,
|
| 253 |
+
"eval_wer": 0.23605024832018698,
|
| 254 |
+
"step": 4400
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 12.53,
|
| 258 |
+
"learning_rate": 9.877557866487755e-05,
|
| 259 |
+
"loss": 0.1069,
|
| 260 |
+
"step": 4500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 12.81,
|
| 264 |
+
"eval_loss": 0.3187991976737976,
|
| 265 |
+
"eval_runtime": 56.3638,
|
| 266 |
+
"eval_samples_per_second": 9.084,
|
| 267 |
+
"eval_steps_per_second": 1.135,
|
| 268 |
+
"eval_wer": 0.2240724510663161,
|
| 269 |
+
"step": 4600
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 13.37,
|
| 273 |
+
"eval_loss": 0.3395535349845886,
|
| 274 |
+
"eval_runtime": 56.3257,
|
| 275 |
+
"eval_samples_per_second": 9.09,
|
| 276 |
+
"eval_steps_per_second": 1.136,
|
| 277 |
+
"eval_wer": 0.21969033011977798,
|
| 278 |
+
"step": 4800
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 13.93,
|
| 282 |
+
"learning_rate": 7.361623616236162e-05,
|
| 283 |
+
"loss": 0.0845,
|
| 284 |
+
"step": 5000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 13.93,
|
| 288 |
+
"eval_loss": 0.3364916741847992,
|
| 289 |
+
"eval_runtime": 56.321,
|
| 290 |
+
"eval_samples_per_second": 9.091,
|
| 291 |
+
"eval_steps_per_second": 1.136,
|
| 292 |
+
"eval_wer": 0.2205667543090856,
|
| 293 |
+
"step": 5000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 14.48,
|
| 297 |
+
"eval_loss": 0.3458584249019623,
|
| 298 |
+
"eval_runtime": 56.3485,
|
| 299 |
+
"eval_samples_per_second": 9.086,
|
| 300 |
+
"eval_steps_per_second": 1.136,
|
| 301 |
+
"eval_wer": 0.22085889570552147,
|
| 302 |
+
"step": 5200
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 15.04,
|
| 306 |
+
"eval_loss": 0.3429270386695862,
|
| 307 |
+
"eval_runtime": 57.9716,
|
| 308 |
+
"eval_samples_per_second": 8.832,
|
| 309 |
+
"eval_steps_per_second": 1.104,
|
| 310 |
+
"eval_wer": 0.2193981887233421,
|
| 311 |
+
"step": 5400
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 15.32,
|
| 315 |
+
"learning_rate": 4.845689365984569e-05,
|
| 316 |
+
"loss": 0.0675,
|
| 317 |
+
"step": 5500
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 15.6,
|
| 321 |
+
"eval_loss": 0.3433798849582672,
|
| 322 |
+
"eval_runtime": 56.3488,
|
| 323 |
+
"eval_samples_per_second": 9.086,
|
| 324 |
+
"eval_steps_per_second": 1.136,
|
| 325 |
+
"eval_wer": 0.2182296231375986,
|
| 326 |
+
"step": 5600
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 16.16,
|
| 330 |
+
"eval_loss": 0.3434172570705414,
|
| 331 |
+
"eval_runtime": 56.3181,
|
| 332 |
+
"eval_samples_per_second": 9.091,
|
| 333 |
+
"eval_steps_per_second": 1.136,
|
| 334 |
+
"eval_wer": 0.20829681565877886,
|
| 335 |
+
"step": 5800
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 16.71,
|
| 339 |
+
"learning_rate": 2.329755115732975e-05,
|
| 340 |
+
"loss": 0.0561,
|
| 341 |
+
"step": 6000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 16.71,
|
| 345 |
+
"eval_loss": 0.33747875690460205,
|
| 346 |
+
"eval_runtime": 56.3763,
|
| 347 |
+
"eval_samples_per_second": 9.082,
|
| 348 |
+
"eval_steps_per_second": 1.135,
|
| 349 |
+
"eval_wer": 0.20362255331580484,
|
| 350 |
+
"step": 6000
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 17.27,
|
| 354 |
+
"eval_loss": 0.3445747494697571,
|
| 355 |
+
"eval_runtime": 56.6382,
|
| 356 |
+
"eval_samples_per_second": 9.04,
|
| 357 |
+
"eval_steps_per_second": 1.13,
|
| 358 |
+
"eval_wer": 0.19865614957639496,
|
| 359 |
+
"step": 6200
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 17.83,
|
| 363 |
+
"eval_loss": 0.33622780442237854,
|
| 364 |
+
"eval_runtime": 56.3321,
|
| 365 |
+
"eval_samples_per_second": 9.089,
|
| 366 |
+
"eval_steps_per_second": 1.136,
|
| 367 |
+
"eval_wer": 0.19777972538708735,
|
| 368 |
+
"step": 6400
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 18.0,
|
| 372 |
+
"step": 6462,
|
| 373 |
+
"total_flos": 3.4883692787272507e+19,
|
| 374 |
+
"train_loss": 0.4896983008235647,
|
| 375 |
+
"train_runtime": 62213.9785,
|
| 376 |
+
"train_samples_per_second": 2.494,
|
| 377 |
+
"train_steps_per_second": 0.104
|
| 378 |
+
}
|
| 379 |
+
],
|
| 380 |
+
"max_steps": 6462,
|
| 381 |
+
"num_train_epochs": 18,
|
| 382 |
+
"total_flos": 3.4883692787272507e+19,
|
| 383 |
+
"trial_name": null,
|
| 384 |
+
"trial_params": null
|
| 385 |
+
}
|