Commit
·
8ee19cb
1
Parent(s):
02d3a9c
End of training
Browse files- all_results.json +16 -0
- eval_results.json +10 -0
- train_results.json +9 -0
- trainer_state.json +319 -0
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.0,
|
| 3 |
+
"eval_cer": 0.9618562329390354,
|
| 4 |
+
"eval_loss": 9.083955764770508,
|
| 5 |
+
"eval_runtime": 52.7023,
|
| 6 |
+
"eval_samples": 542,
|
| 7 |
+
"eval_samples_per_second": 10.284,
|
| 8 |
+
"eval_steps_per_second": 1.29,
|
| 9 |
+
"eval_wer": 1.0,
|
| 10 |
+
"total_flos": 3.373756667888732e+18,
|
| 11 |
+
"train_loss": 5.613170018271794,
|
| 12 |
+
"train_runtime": 5412.6066,
|
| 13 |
+
"train_samples": 448,
|
| 14 |
+
"train_samples_per_second": 12.415,
|
| 15 |
+
"train_steps_per_second": 0.388
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.0,
|
| 3 |
+
"eval_cer": 0.9618562329390354,
|
| 4 |
+
"eval_loss": 9.083955764770508,
|
| 5 |
+
"eval_runtime": 52.7023,
|
| 6 |
+
"eval_samples": 542,
|
| 7 |
+
"eval_samples_per_second": 10.284,
|
| 8 |
+
"eval_steps_per_second": 1.29,
|
| 9 |
+
"eval_wer": 1.0
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.0,
|
| 3 |
+
"total_flos": 3.373756667888732e+18,
|
| 4 |
+
"train_loss": 5.613170018271794,
|
| 5 |
+
"train_runtime": 5412.6066,
|
| 6 |
+
"train_samples": 448,
|
| 7 |
+
"train_samples_per_second": 12.415,
|
| 8 |
+
"train_steps_per_second": 0.388
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,319 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 9.083955764770508,
|
| 3 |
+
"best_model_checkpoint": "wav2vec2-large-xlsr-coraa-exp-2/checkpoint-98",
|
| 4 |
+
"epoch": 27.0,
|
| 5 |
+
"global_step": 378,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.07,
|
| 12 |
+
"learning_rate": 3e-05,
|
| 13 |
+
"loss": 38.902,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_cer": 0.958252957233849,
|
| 19 |
+
"eval_loss": 30.994285583496094,
|
| 20 |
+
"eval_runtime": 52.776,
|
| 21 |
+
"eval_samples_per_second": 10.27,
|
| 22 |
+
"eval_steps_per_second": 1.288,
|
| 23 |
+
"eval_wer": 0.9993904916700528,
|
| 24 |
+
"step": 14
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 2.0,
|
| 28 |
+
"eval_cer": 0.9618562329390354,
|
| 29 |
+
"eval_loss": 13.144615173339844,
|
| 30 |
+
"eval_runtime": 52.6102,
|
| 31 |
+
"eval_samples_per_second": 10.302,
|
| 32 |
+
"eval_steps_per_second": 1.293,
|
| 33 |
+
"eval_wer": 1.0,
|
| 34 |
+
"step": 28
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 3.0,
|
| 38 |
+
"eval_cer": 0.9618562329390354,
|
| 39 |
+
"eval_loss": 10.957932472229004,
|
| 40 |
+
"eval_runtime": 52.5751,
|
| 41 |
+
"eval_samples_per_second": 10.309,
|
| 42 |
+
"eval_steps_per_second": 1.293,
|
| 43 |
+
"eval_wer": 1.0,
|
| 44 |
+
"step": 42
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 4.0,
|
| 48 |
+
"eval_cer": 0.9618562329390354,
|
| 49 |
+
"eval_loss": 10.393542289733887,
|
| 50 |
+
"eval_runtime": 52.6054,
|
| 51 |
+
"eval_samples_per_second": 10.303,
|
| 52 |
+
"eval_steps_per_second": 1.293,
|
| 53 |
+
"eval_wer": 1.0,
|
| 54 |
+
"step": 56
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 5.0,
|
| 58 |
+
"eval_cer": 0.9618562329390354,
|
| 59 |
+
"eval_loss": 9.920334815979004,
|
| 60 |
+
"eval_runtime": 52.6799,
|
| 61 |
+
"eval_samples_per_second": 10.289,
|
| 62 |
+
"eval_steps_per_second": 1.291,
|
| 63 |
+
"eval_wer": 1.0,
|
| 64 |
+
"step": 70
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 6.0,
|
| 68 |
+
"eval_cer": 0.9618562329390354,
|
| 69 |
+
"eval_loss": 9.418496131896973,
|
| 70 |
+
"eval_runtime": 52.6015,
|
| 71 |
+
"eval_samples_per_second": 10.304,
|
| 72 |
+
"eval_steps_per_second": 1.293,
|
| 73 |
+
"eval_wer": 1.0,
|
| 74 |
+
"step": 84
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 7.0,
|
| 78 |
+
"eval_cer": 0.9618562329390354,
|
| 79 |
+
"eval_loss": 9.083955764770508,
|
| 80 |
+
"eval_runtime": 52.5896,
|
| 81 |
+
"eval_samples_per_second": 10.306,
|
| 82 |
+
"eval_steps_per_second": 1.293,
|
| 83 |
+
"eval_wer": 1.0,
|
| 84 |
+
"step": 98
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 7.14,
|
| 88 |
+
"learning_rate": 2.8614285714285714e-05,
|
| 89 |
+
"loss": 11.1539,
|
| 90 |
+
"step": 100
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 8.0,
|
| 94 |
+
"eval_cer": 0.9610191082802548,
|
| 95 |
+
"eval_loss": 9.091231346130371,
|
| 96 |
+
"eval_runtime": 52.5466,
|
| 97 |
+
"eval_samples_per_second": 10.315,
|
| 98 |
+
"eval_steps_per_second": 1.294,
|
| 99 |
+
"eval_wer": 0.9993904916700528,
|
| 100 |
+
"step": 112
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 9.0,
|
| 104 |
+
"eval_cer": 0.913739763421292,
|
| 105 |
+
"eval_loss": 9.123886108398438,
|
| 106 |
+
"eval_runtime": 52.5882,
|
| 107 |
+
"eval_samples_per_second": 10.307,
|
| 108 |
+
"eval_steps_per_second": 1.293,
|
| 109 |
+
"eval_wer": 0.9849654611946363,
|
| 110 |
+
"step": 126
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 10.0,
|
| 114 |
+
"eval_cer": 0.9035122838944495,
|
| 115 |
+
"eval_loss": 9.131282806396484,
|
| 116 |
+
"eval_runtime": 52.5698,
|
| 117 |
+
"eval_samples_per_second": 10.31,
|
| 118 |
+
"eval_steps_per_second": 1.294,
|
| 119 |
+
"eval_wer": 0.9809020723283218,
|
| 120 |
+
"step": 140
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 11.0,
|
| 124 |
+
"eval_cer": 0.8705004549590537,
|
| 125 |
+
"eval_loss": 9.210026741027832,
|
| 126 |
+
"eval_runtime": 52.6067,
|
| 127 |
+
"eval_samples_per_second": 10.303,
|
| 128 |
+
"eval_steps_per_second": 1.293,
|
| 129 |
+
"eval_wer": 0.9867939861844779,
|
| 130 |
+
"step": 154
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 12.0,
|
| 134 |
+
"eval_cer": 0.8711191992720655,
|
| 135 |
+
"eval_loss": 9.104254722595215,
|
| 136 |
+
"eval_runtime": 52.6832,
|
| 137 |
+
"eval_samples_per_second": 10.288,
|
| 138 |
+
"eval_steps_per_second": 1.291,
|
| 139 |
+
"eval_wer": 0.982933766761479,
|
| 140 |
+
"step": 168
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 13.0,
|
| 144 |
+
"eval_cer": 0.9264058234758872,
|
| 145 |
+
"eval_loss": 9.461697578430176,
|
| 146 |
+
"eval_runtime": 52.6083,
|
| 147 |
+
"eval_samples_per_second": 10.303,
|
| 148 |
+
"eval_steps_per_second": 1.293,
|
| 149 |
+
"eval_wer": 0.9792767167817961,
|
| 150 |
+
"step": 182
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 14.0,
|
| 154 |
+
"eval_cer": 0.9234940855323021,
|
| 155 |
+
"eval_loss": 9.622843742370605,
|
| 156 |
+
"eval_runtime": 52.5927,
|
| 157 |
+
"eval_samples_per_second": 10.306,
|
| 158 |
+
"eval_steps_per_second": 1.293,
|
| 159 |
+
"eval_wer": 0.9782608695652174,
|
| 160 |
+
"step": 196
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 14.29,
|
| 164 |
+
"learning_rate": 2.7185714285714287e-05,
|
| 165 |
+
"loss": 3.9522,
|
| 166 |
+
"step": 200
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 15.0,
|
| 170 |
+
"eval_cer": 0.8736669699727024,
|
| 171 |
+
"eval_loss": 9.32619857788086,
|
| 172 |
+
"eval_runtime": 52.6754,
|
| 173 |
+
"eval_samples_per_second": 10.289,
|
| 174 |
+
"eval_steps_per_second": 1.291,
|
| 175 |
+
"eval_wer": 0.9774481917919545,
|
| 176 |
+
"step": 210
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 16.0,
|
| 180 |
+
"eval_cer": 0.877270245677889,
|
| 181 |
+
"eval_loss": 9.436551094055176,
|
| 182 |
+
"eval_runtime": 52.7331,
|
| 183 |
+
"eval_samples_per_second": 10.278,
|
| 184 |
+
"eval_steps_per_second": 1.29,
|
| 185 |
+
"eval_wer": 0.9764323445753759,
|
| 186 |
+
"step": 224
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 17.0,
|
| 190 |
+
"eval_cer": 0.90049135577798,
|
| 191 |
+
"eval_loss": 9.723479270935059,
|
| 192 |
+
"eval_runtime": 52.6226,
|
| 193 |
+
"eval_samples_per_second": 10.3,
|
| 194 |
+
"eval_steps_per_second": 1.292,
|
| 195 |
+
"eval_wer": 0.9758228362454286,
|
| 196 |
+
"step": 238
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 18.0,
|
| 200 |
+
"eval_cer": 0.8971064604185623,
|
| 201 |
+
"eval_loss": 9.901185989379883,
|
| 202 |
+
"eval_runtime": 52.6533,
|
| 203 |
+
"eval_samples_per_second": 10.294,
|
| 204 |
+
"eval_steps_per_second": 1.291,
|
| 205 |
+
"eval_wer": 0.9750101584721658,
|
| 206 |
+
"step": 252
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 19.0,
|
| 210 |
+
"eval_cer": 0.8567060964513193,
|
| 211 |
+
"eval_loss": 9.75187873840332,
|
| 212 |
+
"eval_runtime": 52.6716,
|
| 213 |
+
"eval_samples_per_second": 10.29,
|
| 214 |
+
"eval_steps_per_second": 1.291,
|
| 215 |
+
"eval_wer": 0.9845591223080049,
|
| 216 |
+
"step": 266
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 20.0,
|
| 220 |
+
"eval_cer": 0.8975796178343949,
|
| 221 |
+
"eval_loss": 9.862020492553711,
|
| 222 |
+
"eval_runtime": 52.6667,
|
| 223 |
+
"eval_samples_per_second": 10.291,
|
| 224 |
+
"eval_steps_per_second": 1.291,
|
| 225 |
+
"eval_wer": 0.9764323445753759,
|
| 226 |
+
"step": 280
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 21.0,
|
| 230 |
+
"eval_cer": 0.8681346678798908,
|
| 231 |
+
"eval_loss": 9.89448070526123,
|
| 232 |
+
"eval_runtime": 52.6476,
|
| 233 |
+
"eval_samples_per_second": 10.295,
|
| 234 |
+
"eval_steps_per_second": 1.292,
|
| 235 |
+
"eval_wer": 0.9782608695652174,
|
| 236 |
+
"step": 294
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 21.43,
|
| 240 |
+
"learning_rate": 2.5757142857142857e-05,
|
| 241 |
+
"loss": 3.3723,
|
| 242 |
+
"step": 300
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 22.0,
|
| 246 |
+
"eval_cer": 0.8705732484076433,
|
| 247 |
+
"eval_loss": 9.991724967956543,
|
| 248 |
+
"eval_runtime": 52.601,
|
| 249 |
+
"eval_samples_per_second": 10.304,
|
| 250 |
+
"eval_steps_per_second": 1.293,
|
| 251 |
+
"eval_wer": 0.9748069890288501,
|
| 252 |
+
"step": 308
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 23.0,
|
| 256 |
+
"eval_cer": 0.8577252047315742,
|
| 257 |
+
"eval_loss": 10.057960510253906,
|
| 258 |
+
"eval_runtime": 52.6905,
|
| 259 |
+
"eval_samples_per_second": 10.286,
|
| 260 |
+
"eval_steps_per_second": 1.291,
|
| 261 |
+
"eval_wer": 0.980089394555059,
|
| 262 |
+
"step": 322
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 24.0,
|
| 266 |
+
"eval_cer": 0.8707552320291174,
|
| 267 |
+
"eval_loss": 10.1575345993042,
|
| 268 |
+
"eval_runtime": 52.7752,
|
| 269 |
+
"eval_samples_per_second": 10.27,
|
| 270 |
+
"eval_steps_per_second": 1.288,
|
| 271 |
+
"eval_wer": 0.9764323445753759,
|
| 272 |
+
"step": 336
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 25.0,
|
| 276 |
+
"eval_cer": 0.8554686078252958,
|
| 277 |
+
"eval_loss": 10.198201179504395,
|
| 278 |
+
"eval_runtime": 52.726,
|
| 279 |
+
"eval_samples_per_second": 10.28,
|
| 280 |
+
"eval_steps_per_second": 1.29,
|
| 281 |
+
"eval_wer": 0.9819179195449005,
|
| 282 |
+
"step": 350
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 26.0,
|
| 286 |
+
"eval_cer": 0.9087170154686078,
|
| 287 |
+
"eval_loss": 10.40873908996582,
|
| 288 |
+
"eval_runtime": 53.2414,
|
| 289 |
+
"eval_samples_per_second": 10.18,
|
| 290 |
+
"eval_steps_per_second": 1.277,
|
| 291 |
+
"eval_wer": 0.9790735473384803,
|
| 292 |
+
"step": 364
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 27.0,
|
| 296 |
+
"eval_cer": 0.8901910828025478,
|
| 297 |
+
"eval_loss": 10.506359100341797,
|
| 298 |
+
"eval_runtime": 52.6529,
|
| 299 |
+
"eval_samples_per_second": 10.294,
|
| 300 |
+
"eval_steps_per_second": 1.291,
|
| 301 |
+
"eval_wer": 0.9798862251117432,
|
| 302 |
+
"step": 378
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 27.0,
|
| 306 |
+
"step": 378,
|
| 307 |
+
"total_flos": 3.373756667888732e+18,
|
| 308 |
+
"train_loss": 5.613170018271794,
|
| 309 |
+
"train_runtime": 5412.6066,
|
| 310 |
+
"train_samples_per_second": 12.415,
|
| 311 |
+
"train_steps_per_second": 0.388
|
| 312 |
+
}
|
| 313 |
+
],
|
| 314 |
+
"max_steps": 2100,
|
| 315 |
+
"num_train_epochs": 150,
|
| 316 |
+
"total_flos": 3.373756667888732e+18,
|
| 317 |
+
"trial_name": null,
|
| 318 |
+
"trial_params": null
|
| 319 |
+
}
|