End of training
Browse files- .ipynb_checkpoints/trainer_state-checkpoint.json +442 -0
- README.md +54 -0
- config.json +117 -0
- model.safetensors +3 -0
- trainer_state.json +442 -0
- training_args.bin +3 -0
.ipynb_checkpoints/trainer_state-checkpoint.json
ADDED
|
@@ -0,0 +1,442 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.24260137975215912,
|
| 3 |
+
"best_model_checkpoint": "wave2vec2-xlsr-Persian/checkpoint-240000",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 10000,
|
| 6 |
+
"global_step": 255012,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1176415227518705,
|
| 13 |
+
"grad_norm": 1.416466236114502,
|
| 14 |
+
"learning_rate": 9.646079712769477e-06,
|
| 15 |
+
"loss": 3.7691,
|
| 16 |
+
"step": 10000
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1176415227518705,
|
| 20 |
+
"eval_loss": 0.7940966486930847,
|
| 21 |
+
"eval_runtime": 534.4277,
|
| 22 |
+
"eval_samples_per_second": 16.743,
|
| 23 |
+
"eval_steps_per_second": 2.094,
|
| 24 |
+
"eval_wer": 0.6079105283797882,
|
| 25 |
+
"step": 10000
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.235283045503741,
|
| 29 |
+
"grad_norm": 2.3572049140930176,
|
| 30 |
+
"learning_rate": 9.252436892745226e-06,
|
| 31 |
+
"loss": 0.8658,
|
| 32 |
+
"step": 20000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.235283045503741,
|
| 36 |
+
"eval_loss": 0.5118501782417297,
|
| 37 |
+
"eval_runtime": 536.8454,
|
| 38 |
+
"eval_samples_per_second": 16.668,
|
| 39 |
+
"eval_steps_per_second": 2.084,
|
| 40 |
+
"eval_wer": 0.4575246579700923,
|
| 41 |
+
"step": 20000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.3529245682556115,
|
| 45 |
+
"grad_norm": 2.209596872329712,
|
| 46 |
+
"learning_rate": 8.858872809158622e-06,
|
| 47 |
+
"loss": 0.6829,
|
| 48 |
+
"step": 30000
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.3529245682556115,
|
| 52 |
+
"eval_loss": 0.4285117983818054,
|
| 53 |
+
"eval_runtime": 539.6896,
|
| 54 |
+
"eval_samples_per_second": 16.58,
|
| 55 |
+
"eval_steps_per_second": 2.073,
|
| 56 |
+
"eval_wer": 0.40392825696083584,
|
| 57 |
+
"step": 30000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.470566091007482,
|
| 61 |
+
"grad_norm": 2.2252326011657715,
|
| 62 |
+
"learning_rate": 8.465348093790845e-06,
|
| 63 |
+
"loss": 0.6078,
|
| 64 |
+
"step": 40000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.470566091007482,
|
| 68 |
+
"eval_loss": 0.38793477416038513,
|
| 69 |
+
"eval_runtime": 547.0046,
|
| 70 |
+
"eval_samples_per_second": 16.358,
|
| 71 |
+
"eval_steps_per_second": 2.046,
|
| 72 |
+
"eval_wer": 0.3664437028760849,
|
| 73 |
+
"step": 40000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.5882076137593525,
|
| 77 |
+
"grad_norm": 2.7398128509521484,
|
| 78 |
+
"learning_rate": 8.071784010204244e-06,
|
| 79 |
+
"loss": 0.5561,
|
| 80 |
+
"step": 50000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.5882076137593525,
|
| 84 |
+
"eval_loss": 0.3593791723251343,
|
| 85 |
+
"eval_runtime": 551.125,
|
| 86 |
+
"eval_samples_per_second": 16.236,
|
| 87 |
+
"eval_steps_per_second": 2.03,
|
| 88 |
+
"eval_wer": 0.34282162650112097,
|
| 89 |
+
"step": 50000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.705849136511223,
|
| 93 |
+
"grad_norm": 2.960217237472534,
|
| 94 |
+
"learning_rate": 7.678259294836465e-06,
|
| 95 |
+
"loss": 0.5168,
|
| 96 |
+
"step": 60000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.705849136511223,
|
| 100 |
+
"eval_loss": 0.33377397060394287,
|
| 101 |
+
"eval_runtime": 546.4831,
|
| 102 |
+
"eval_samples_per_second": 16.374,
|
| 103 |
+
"eval_steps_per_second": 2.048,
|
| 104 |
+
"eval_wer": 0.3202835389089079,
|
| 105 |
+
"step": 60000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.8234906592630935,
|
| 109 |
+
"grad_norm": 2.531191349029541,
|
| 110 |
+
"learning_rate": 7.284734579468687e-06,
|
| 111 |
+
"loss": 0.499,
|
| 112 |
+
"step": 70000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.8234906592630935,
|
| 116 |
+
"eval_loss": 0.3143016993999481,
|
| 117 |
+
"eval_runtime": 550.3802,
|
| 118 |
+
"eval_samples_per_second": 16.258,
|
| 119 |
+
"eval_steps_per_second": 2.033,
|
| 120 |
+
"eval_wer": 0.30924387157877603,
|
| 121 |
+
"step": 70000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.941132182014964,
|
| 125 |
+
"grad_norm": 2.0980894565582275,
|
| 126 |
+
"learning_rate": 6.891131127663261e-06,
|
| 127 |
+
"loss": 0.4768,
|
| 128 |
+
"step": 80000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.941132182014964,
|
| 132 |
+
"eval_loss": 0.3023754954338074,
|
| 133 |
+
"eval_runtime": 546.9538,
|
| 134 |
+
"eval_samples_per_second": 16.36,
|
| 135 |
+
"eval_steps_per_second": 2.046,
|
| 136 |
+
"eval_wer": 0.29365367113334173,
|
| 137 |
+
"step": 80000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.0587737047668344,
|
| 141 |
+
"grad_norm": 1.7364046573638916,
|
| 142 |
+
"learning_rate": 6.497606412295483e-06,
|
| 143 |
+
"loss": 0.4529,
|
| 144 |
+
"step": 90000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.0587737047668344,
|
| 148 |
+
"eval_loss": 0.29402047395706177,
|
| 149 |
+
"eval_runtime": 544.3376,
|
| 150 |
+
"eval_samples_per_second": 16.438,
|
| 151 |
+
"eval_steps_per_second": 2.056,
|
| 152 |
+
"eval_wer": 0.2827878859629002,
|
| 153 |
+
"step": 90000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.176415227518705,
|
| 157 |
+
"grad_norm": 3.4313299655914307,
|
| 158 |
+
"learning_rate": 6.1040816969277054e-06,
|
| 159 |
+
"loss": 0.44,
|
| 160 |
+
"step": 100000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.176415227518705,
|
| 164 |
+
"eval_loss": 0.2909528613090515,
|
| 165 |
+
"eval_runtime": 543.2034,
|
| 166 |
+
"eval_samples_per_second": 16.473,
|
| 167 |
+
"eval_steps_per_second": 2.06,
|
| 168 |
+
"eval_wer": 0.27463762217996435,
|
| 169 |
+
"step": 100000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.2940567502705755,
|
| 173 |
+
"grad_norm": 1.1621043682098389,
|
| 174 |
+
"learning_rate": 5.710478245122278e-06,
|
| 175 |
+
"loss": 0.4264,
|
| 176 |
+
"step": 110000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.2940567502705755,
|
| 180 |
+
"eval_loss": 0.2814837098121643,
|
| 181 |
+
"eval_runtime": 547.8907,
|
| 182 |
+
"eval_samples_per_second": 16.332,
|
| 183 |
+
"eval_steps_per_second": 2.042,
|
| 184 |
+
"eval_wer": 0.2683001724023115,
|
| 185 |
+
"step": 110000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.4116982730224459,
|
| 189 |
+
"grad_norm": 2.507716178894043,
|
| 190 |
+
"learning_rate": 5.316914161535675e-06,
|
| 191 |
+
"loss": 0.4189,
|
| 192 |
+
"step": 120000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.4116982730224459,
|
| 196 |
+
"eval_loss": 0.27175650000572205,
|
| 197 |
+
"eval_runtime": 548.1953,
|
| 198 |
+
"eval_samples_per_second": 16.323,
|
| 199 |
+
"eval_steps_per_second": 2.041,
|
| 200 |
+
"eval_wer": 0.26368304611946813,
|
| 201 |
+
"step": 120000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.5293397957743164,
|
| 205 |
+
"grad_norm": 2.3356528282165527,
|
| 206 |
+
"learning_rate": 4.923428814386722e-06,
|
| 207 |
+
"loss": 0.4052,
|
| 208 |
+
"step": 130000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.5293397957743164,
|
| 212 |
+
"eval_loss": 0.2673029899597168,
|
| 213 |
+
"eval_runtime": 544.6822,
|
| 214 |
+
"eval_samples_per_second": 16.428,
|
| 215 |
+
"eval_steps_per_second": 2.054,
|
| 216 |
+
"eval_wer": 0.25851467639420195,
|
| 217 |
+
"step": 130000
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 1.646981318526187,
|
| 221 |
+
"grad_norm": 2.476557970046997,
|
| 222 |
+
"learning_rate": 4.529864730800119e-06,
|
| 223 |
+
"loss": 0.4044,
|
| 224 |
+
"step": 140000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.646981318526187,
|
| 228 |
+
"eval_loss": 0.26591917872428894,
|
| 229 |
+
"eval_runtime": 545.1182,
|
| 230 |
+
"eval_samples_per_second": 16.415,
|
| 231 |
+
"eval_steps_per_second": 2.053,
|
| 232 |
+
"eval_wer": 0.2534720937631799,
|
| 233 |
+
"step": 140000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.7646228412780576,
|
| 237 |
+
"grad_norm": 1.8470633029937744,
|
| 238 |
+
"learning_rate": 4.136418751869991e-06,
|
| 239 |
+
"loss": 0.4046,
|
| 240 |
+
"step": 150000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.7646228412780576,
|
| 244 |
+
"eval_loss": 0.2603091299533844,
|
| 245 |
+
"eval_runtime": 545.5751,
|
| 246 |
+
"eval_samples_per_second": 16.401,
|
| 247 |
+
"eval_steps_per_second": 2.051,
|
| 248 |
+
"eval_wer": 0.24952459877616556,
|
| 249 |
+
"step": 150000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.8822643640299281,
|
| 253 |
+
"grad_norm": 2.6165308952331543,
|
| 254 |
+
"learning_rate": 3.742815300064564e-06,
|
| 255 |
+
"loss": 0.3944,
|
| 256 |
+
"step": 160000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.8822643640299281,
|
| 260 |
+
"eval_loss": 0.2575734555721283,
|
| 261 |
+
"eval_runtime": 545.6155,
|
| 262 |
+
"eval_samples_per_second": 16.4,
|
| 263 |
+
"eval_steps_per_second": 2.051,
|
| 264 |
+
"eval_wer": 0.24606545368445198,
|
| 265 |
+
"step": 160000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 1.9999058867817985,
|
| 269 |
+
"grad_norm": 2.0350422859191895,
|
| 270 |
+
"learning_rate": 3.3492512164779615e-06,
|
| 271 |
+
"loss": 0.3876,
|
| 272 |
+
"step": 170000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.9999058867817985,
|
| 276 |
+
"eval_loss": 0.25538763403892517,
|
| 277 |
+
"eval_runtime": 543.8449,
|
| 278 |
+
"eval_samples_per_second": 16.453,
|
| 279 |
+
"eval_steps_per_second": 2.058,
|
| 280 |
+
"eval_wer": 0.24299106911630866,
|
| 281 |
+
"step": 170000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.117547409533669,
|
| 285 |
+
"grad_norm": 2.234062433242798,
|
| 286 |
+
"learning_rate": 2.955726501110184e-06,
|
| 287 |
+
"loss": 0.3836,
|
| 288 |
+
"step": 180000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.117547409533669,
|
| 292 |
+
"eval_loss": 0.25172147154808044,
|
| 293 |
+
"eval_runtime": 543.9926,
|
| 294 |
+
"eval_samples_per_second": 16.449,
|
| 295 |
+
"eval_steps_per_second": 2.057,
|
| 296 |
+
"eval_wer": 0.24226224389377649,
|
| 297 |
+
"step": 180000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 2.2351889322855394,
|
| 301 |
+
"grad_norm": 2.7782626152038574,
|
| 302 |
+
"learning_rate": 2.5621624175235817e-06,
|
| 303 |
+
"loss": 0.3767,
|
| 304 |
+
"step": 190000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 2.2351889322855394,
|
| 308 |
+
"eval_loss": 0.2503082752227783,
|
| 309 |
+
"eval_runtime": 546.7703,
|
| 310 |
+
"eval_samples_per_second": 16.365,
|
| 311 |
+
"eval_steps_per_second": 2.047,
|
| 312 |
+
"eval_wer": 0.23942833465286462,
|
| 313 |
+
"step": 190000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 2.35283045503741,
|
| 317 |
+
"grad_norm": 2.365490198135376,
|
| 318 |
+
"learning_rate": 2.1686770703746284e-06,
|
| 319 |
+
"loss": 0.3738,
|
| 320 |
+
"step": 200000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.35283045503741,
|
| 324 |
+
"eval_loss": 0.24804162979125977,
|
| 325 |
+
"eval_runtime": 545.3342,
|
| 326 |
+
"eval_samples_per_second": 16.408,
|
| 327 |
+
"eval_steps_per_second": 2.052,
|
| 328 |
+
"eval_wer": 0.23909166919474062,
|
| 329 |
+
"step": 200000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.4704719777892805,
|
| 333 |
+
"grad_norm": 3.105099678039551,
|
| 334 |
+
"learning_rate": 1.7751523550068502e-06,
|
| 335 |
+
"loss": 0.3707,
|
| 336 |
+
"step": 210000
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 2.4704719777892805,
|
| 340 |
+
"eval_loss": 0.24506914615631104,
|
| 341 |
+
"eval_runtime": 547.7088,
|
| 342 |
+
"eval_samples_per_second": 16.337,
|
| 343 |
+
"eval_steps_per_second": 2.043,
|
| 344 |
+
"eval_wer": 0.23790409104025928,
|
| 345 |
+
"step": 210000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.588113500541151,
|
| 349 |
+
"grad_norm": 0.9898041486740112,
|
| 350 |
+
"learning_rate": 1.3815882714202481e-06,
|
| 351 |
+
"loss": 0.3649,
|
| 352 |
+
"step": 220000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 2.588113500541151,
|
| 356 |
+
"eval_loss": 0.24370211362838745,
|
| 357 |
+
"eval_runtime": 547.4398,
|
| 358 |
+
"eval_samples_per_second": 16.345,
|
| 359 |
+
"eval_steps_per_second": 2.044,
|
| 360 |
+
"eval_wer": 0.23637614780723498,
|
| 361 |
+
"step": 220000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.7057550232930216,
|
| 365 |
+
"grad_norm": 4.53593635559082,
|
| 366 |
+
"learning_rate": 9.880241878336459e-07,
|
| 367 |
+
"loss": 0.369,
|
| 368 |
+
"step": 230000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.7057550232930216,
|
| 372 |
+
"eval_loss": 0.24421393871307373,
|
| 373 |
+
"eval_runtime": 549.6273,
|
| 374 |
+
"eval_samples_per_second": 16.28,
|
| 375 |
+
"eval_steps_per_second": 2.036,
|
| 376 |
+
"eval_wer": 0.23538094991453876,
|
| 377 |
+
"step": 230000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.8233965460448918,
|
| 381 |
+
"grad_norm": 3.995215654373169,
|
| 382 |
+
"learning_rate": 5.945388406846922e-07,
|
| 383 |
+
"loss": 0.3608,
|
| 384 |
+
"step": 240000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.8233965460448918,
|
| 388 |
+
"eval_loss": 0.24260137975215912,
|
| 389 |
+
"eval_runtime": 548.4399,
|
| 390 |
+
"eval_samples_per_second": 16.315,
|
| 391 |
+
"eval_steps_per_second": 2.04,
|
| 392 |
+
"eval_wer": 0.23452633759776248,
|
| 393 |
+
"step": 240000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.9410380687967628,
|
| 397 |
+
"grad_norm": 1.9774836301803589,
|
| 398 |
+
"learning_rate": 2.0097475709808989e-07,
|
| 399 |
+
"loss": 0.3644,
|
| 400 |
+
"step": 250000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 2.9410380687967628,
|
| 404 |
+
"eval_loss": 0.24260272085666656,
|
| 405 |
+
"eval_runtime": 548.9964,
|
| 406 |
+
"eval_samples_per_second": 16.299,
|
| 407 |
+
"eval_steps_per_second": 2.038,
|
| 408 |
+
"eval_wer": 0.234019489600367,
|
| 409 |
+
"step": 250000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 3.0,
|
| 413 |
+
"step": 255012,
|
| 414 |
+
"total_flos": 2.7896646019563717e+20,
|
| 415 |
+
"train_loss": 0.5826510110653187,
|
| 416 |
+
"train_runtime": 84804.3792,
|
| 417 |
+
"train_samples_per_second": 6.014,
|
| 418 |
+
"train_steps_per_second": 3.007
|
| 419 |
+
}
|
| 420 |
+
],
|
| 421 |
+
"logging_steps": 10000,
|
| 422 |
+
"max_steps": 255012,
|
| 423 |
+
"num_input_tokens_seen": 0,
|
| 424 |
+
"num_train_epochs": 3,
|
| 425 |
+
"save_steps": 10000,
|
| 426 |
+
"stateful_callbacks": {
|
| 427 |
+
"TrainerControl": {
|
| 428 |
+
"args": {
|
| 429 |
+
"should_epoch_stop": false,
|
| 430 |
+
"should_evaluate": false,
|
| 431 |
+
"should_log": false,
|
| 432 |
+
"should_save": true,
|
| 433 |
+
"should_training_stop": true
|
| 434 |
+
},
|
| 435 |
+
"attributes": {}
|
| 436 |
+
}
|
| 437 |
+
},
|
| 438 |
+
"total_flos": 2.7896646019563717e+20,
|
| 439 |
+
"train_batch_size": 2,
|
| 440 |
+
"trial_name": null,
|
| 441 |
+
"trial_params": null
|
| 442 |
+
}
|
README.md
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- generated_from_trainer
|
| 4 |
+
model-index:
|
| 5 |
+
- name: wave2vec2-xlsr-Persian
|
| 6 |
+
results: []
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 10 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 11 |
+
|
| 12 |
+
# wave2vec2-xlsr-Persian
|
| 13 |
+
|
| 14 |
+
This model was trained from scratch on an unknown dataset.
|
| 15 |
+
It achieves the following results on the evaluation set:
|
| 16 |
+
- epoch: 2.9410
|
| 17 |
+
- eval_loss: 0.2426
|
| 18 |
+
- eval_runtime: 548.9964
|
| 19 |
+
- eval_samples_per_second: 16.299
|
| 20 |
+
- eval_steps_per_second: 2.038
|
| 21 |
+
- eval_wer: 0.2340
|
| 22 |
+
- step: 250000
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 5e-05
|
| 42 |
+
- train_batch_size: 8
|
| 43 |
+
- eval_batch_size: 8
|
| 44 |
+
- seed: 42
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: linear
|
| 47 |
+
- num_epochs: 3.0
|
| 48 |
+
|
| 49 |
+
### Framework versions
|
| 50 |
+
|
| 51 |
+
- Transformers 4.44.0
|
| 52 |
+
- Pytorch 2.4.0.dev20240522
|
| 53 |
+
- Datasets 2.19.1
|
| 54 |
+
- Tokenizers 0.19.1
|
config.json
ADDED
|
@@ -0,0 +1,117 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./wave2vec2-xlsr-Persian/checkpoint-255012",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"adapter_attn_dim": null,
|
| 5 |
+
"adapter_kernel_size": 3,
|
| 6 |
+
"adapter_stride": 2,
|
| 7 |
+
"add_adapter": false,
|
| 8 |
+
"apply_spec_augment": true,
|
| 9 |
+
"architectures": [
|
| 10 |
+
"Wav2Vec2ForCTC"
|
| 11 |
+
],
|
| 12 |
+
"attention_dropout": 0.1,
|
| 13 |
+
"bos_token_id": 1,
|
| 14 |
+
"classifier_proj_size": 256,
|
| 15 |
+
"codevector_dim": 768,
|
| 16 |
+
"contrastive_logits_temperature": 0.1,
|
| 17 |
+
"conv_bias": true,
|
| 18 |
+
"conv_dim": [
|
| 19 |
+
512,
|
| 20 |
+
512,
|
| 21 |
+
512,
|
| 22 |
+
512,
|
| 23 |
+
512,
|
| 24 |
+
512,
|
| 25 |
+
512
|
| 26 |
+
],
|
| 27 |
+
"conv_kernel": [
|
| 28 |
+
10,
|
| 29 |
+
3,
|
| 30 |
+
3,
|
| 31 |
+
3,
|
| 32 |
+
3,
|
| 33 |
+
2,
|
| 34 |
+
2
|
| 35 |
+
],
|
| 36 |
+
"conv_stride": [
|
| 37 |
+
5,
|
| 38 |
+
2,
|
| 39 |
+
2,
|
| 40 |
+
2,
|
| 41 |
+
2,
|
| 42 |
+
2,
|
| 43 |
+
2
|
| 44 |
+
],
|
| 45 |
+
"ctc_loss_reduction": "mean",
|
| 46 |
+
"ctc_zero_infinity": false,
|
| 47 |
+
"diversity_loss_weight": 0.1,
|
| 48 |
+
"do_stable_layer_norm": true,
|
| 49 |
+
"eos_token_id": 2,
|
| 50 |
+
"feat_extract_activation": "gelu",
|
| 51 |
+
"feat_extract_dropout": 0.0,
|
| 52 |
+
"feat_extract_norm": "layer",
|
| 53 |
+
"feat_proj_dropout": 0.1,
|
| 54 |
+
"feat_quantizer_dropout": 0.0,
|
| 55 |
+
"final_dropout": 0.0,
|
| 56 |
+
"gradient_checkpointing": false,
|
| 57 |
+
"hidden_act": "gelu",
|
| 58 |
+
"hidden_dropout": 0.1,
|
| 59 |
+
"hidden_size": 1024,
|
| 60 |
+
"initializer_range": 0.02,
|
| 61 |
+
"intermediate_size": 4096,
|
| 62 |
+
"layer_norm_eps": 1e-05,
|
| 63 |
+
"layerdrop": 0.1,
|
| 64 |
+
"mask_channel_length": 10,
|
| 65 |
+
"mask_channel_min_space": 1,
|
| 66 |
+
"mask_channel_other": 0.0,
|
| 67 |
+
"mask_channel_prob": 0.0,
|
| 68 |
+
"mask_channel_selection": "static",
|
| 69 |
+
"mask_feature_length": 10,
|
| 70 |
+
"mask_feature_min_masks": 0,
|
| 71 |
+
"mask_feature_prob": 0.0,
|
| 72 |
+
"mask_time_length": 10,
|
| 73 |
+
"mask_time_min_masks": 2,
|
| 74 |
+
"mask_time_min_space": 1,
|
| 75 |
+
"mask_time_other": 0.0,
|
| 76 |
+
"mask_time_prob": 0.075,
|
| 77 |
+
"mask_time_selection": "static",
|
| 78 |
+
"model_type": "wav2vec2",
|
| 79 |
+
"num_adapter_layers": 3,
|
| 80 |
+
"num_attention_heads": 16,
|
| 81 |
+
"num_codevector_groups": 2,
|
| 82 |
+
"num_codevectors_per_group": 320,
|
| 83 |
+
"num_conv_pos_embedding_groups": 16,
|
| 84 |
+
"num_conv_pos_embeddings": 128,
|
| 85 |
+
"num_feat_extract_layers": 7,
|
| 86 |
+
"num_hidden_layers": 24,
|
| 87 |
+
"num_negatives": 100,
|
| 88 |
+
"output_hidden_size": 1024,
|
| 89 |
+
"pad_token_id": 0,
|
| 90 |
+
"proj_codevector_dim": 768,
|
| 91 |
+
"tdnn_dilation": [
|
| 92 |
+
1,
|
| 93 |
+
2,
|
| 94 |
+
3,
|
| 95 |
+
1,
|
| 96 |
+
1
|
| 97 |
+
],
|
| 98 |
+
"tdnn_dim": [
|
| 99 |
+
512,
|
| 100 |
+
512,
|
| 101 |
+
512,
|
| 102 |
+
512,
|
| 103 |
+
1500
|
| 104 |
+
],
|
| 105 |
+
"tdnn_kernel": [
|
| 106 |
+
5,
|
| 107 |
+
3,
|
| 108 |
+
3,
|
| 109 |
+
1,
|
| 110 |
+
1
|
| 111 |
+
],
|
| 112 |
+
"torch_dtype": "float32",
|
| 113 |
+
"transformers_version": "4.44.0",
|
| 114 |
+
"use_weighted_layer_sum": false,
|
| 115 |
+
"vocab_size": 40,
|
| 116 |
+
"xvector_output_dim": 512
|
| 117 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:83745a27f317bf60c598c88317bf8f940a3731cc01f0ce362e86ca7c2ce2047b
|
| 3 |
+
size 1261971480
|
trainer_state.json
ADDED
|
@@ -0,0 +1,442 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.24260137975215912,
|
| 3 |
+
"best_model_checkpoint": "wave2vec2-xlsr-Persian/checkpoint-240000",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 10000,
|
| 6 |
+
"global_step": 255012,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1176415227518705,
|
| 13 |
+
"grad_norm": 1.416466236114502,
|
| 14 |
+
"learning_rate": 9.646079712769477e-06,
|
| 15 |
+
"loss": 3.7691,
|
| 16 |
+
"step": 10000
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.1176415227518705,
|
| 20 |
+
"eval_loss": 0.7940966486930847,
|
| 21 |
+
"eval_runtime": 534.4277,
|
| 22 |
+
"eval_samples_per_second": 16.743,
|
| 23 |
+
"eval_steps_per_second": 2.094,
|
| 24 |
+
"eval_wer": 0.6079105283797882,
|
| 25 |
+
"step": 10000
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.235283045503741,
|
| 29 |
+
"grad_norm": 2.3572049140930176,
|
| 30 |
+
"learning_rate": 9.252436892745226e-06,
|
| 31 |
+
"loss": 0.8658,
|
| 32 |
+
"step": 20000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.235283045503741,
|
| 36 |
+
"eval_loss": 0.5118501782417297,
|
| 37 |
+
"eval_runtime": 536.8454,
|
| 38 |
+
"eval_samples_per_second": 16.668,
|
| 39 |
+
"eval_steps_per_second": 2.084,
|
| 40 |
+
"eval_wer": 0.4575246579700923,
|
| 41 |
+
"step": 20000
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.3529245682556115,
|
| 45 |
+
"grad_norm": 2.209596872329712,
|
| 46 |
+
"learning_rate": 8.858872809158622e-06,
|
| 47 |
+
"loss": 0.6829,
|
| 48 |
+
"step": 30000
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.3529245682556115,
|
| 52 |
+
"eval_loss": 0.4285117983818054,
|
| 53 |
+
"eval_runtime": 539.6896,
|
| 54 |
+
"eval_samples_per_second": 16.58,
|
| 55 |
+
"eval_steps_per_second": 2.073,
|
| 56 |
+
"eval_wer": 0.40392825696083584,
|
| 57 |
+
"step": 30000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.470566091007482,
|
| 61 |
+
"grad_norm": 2.2252326011657715,
|
| 62 |
+
"learning_rate": 8.465348093790845e-06,
|
| 63 |
+
"loss": 0.6078,
|
| 64 |
+
"step": 40000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.470566091007482,
|
| 68 |
+
"eval_loss": 0.38793477416038513,
|
| 69 |
+
"eval_runtime": 547.0046,
|
| 70 |
+
"eval_samples_per_second": 16.358,
|
| 71 |
+
"eval_steps_per_second": 2.046,
|
| 72 |
+
"eval_wer": 0.3664437028760849,
|
| 73 |
+
"step": 40000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.5882076137593525,
|
| 77 |
+
"grad_norm": 2.7398128509521484,
|
| 78 |
+
"learning_rate": 8.071784010204244e-06,
|
| 79 |
+
"loss": 0.5561,
|
| 80 |
+
"step": 50000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.5882076137593525,
|
| 84 |
+
"eval_loss": 0.3593791723251343,
|
| 85 |
+
"eval_runtime": 551.125,
|
| 86 |
+
"eval_samples_per_second": 16.236,
|
| 87 |
+
"eval_steps_per_second": 2.03,
|
| 88 |
+
"eval_wer": 0.34282162650112097,
|
| 89 |
+
"step": 50000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.705849136511223,
|
| 93 |
+
"grad_norm": 2.960217237472534,
|
| 94 |
+
"learning_rate": 7.678259294836465e-06,
|
| 95 |
+
"loss": 0.5168,
|
| 96 |
+
"step": 60000
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.705849136511223,
|
| 100 |
+
"eval_loss": 0.33377397060394287,
|
| 101 |
+
"eval_runtime": 546.4831,
|
| 102 |
+
"eval_samples_per_second": 16.374,
|
| 103 |
+
"eval_steps_per_second": 2.048,
|
| 104 |
+
"eval_wer": 0.3202835389089079,
|
| 105 |
+
"step": 60000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.8234906592630935,
|
| 109 |
+
"grad_norm": 2.531191349029541,
|
| 110 |
+
"learning_rate": 7.284734579468687e-06,
|
| 111 |
+
"loss": 0.499,
|
| 112 |
+
"step": 70000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.8234906592630935,
|
| 116 |
+
"eval_loss": 0.3143016993999481,
|
| 117 |
+
"eval_runtime": 550.3802,
|
| 118 |
+
"eval_samples_per_second": 16.258,
|
| 119 |
+
"eval_steps_per_second": 2.033,
|
| 120 |
+
"eval_wer": 0.30924387157877603,
|
| 121 |
+
"step": 70000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.941132182014964,
|
| 125 |
+
"grad_norm": 2.0980894565582275,
|
| 126 |
+
"learning_rate": 6.891131127663261e-06,
|
| 127 |
+
"loss": 0.4768,
|
| 128 |
+
"step": 80000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.941132182014964,
|
| 132 |
+
"eval_loss": 0.3023754954338074,
|
| 133 |
+
"eval_runtime": 546.9538,
|
| 134 |
+
"eval_samples_per_second": 16.36,
|
| 135 |
+
"eval_steps_per_second": 2.046,
|
| 136 |
+
"eval_wer": 0.29365367113334173,
|
| 137 |
+
"step": 80000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.0587737047668344,
|
| 141 |
+
"grad_norm": 1.7364046573638916,
|
| 142 |
+
"learning_rate": 6.497606412295483e-06,
|
| 143 |
+
"loss": 0.4529,
|
| 144 |
+
"step": 90000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.0587737047668344,
|
| 148 |
+
"eval_loss": 0.29402047395706177,
|
| 149 |
+
"eval_runtime": 544.3376,
|
| 150 |
+
"eval_samples_per_second": 16.438,
|
| 151 |
+
"eval_steps_per_second": 2.056,
|
| 152 |
+
"eval_wer": 0.2827878859629002,
|
| 153 |
+
"step": 90000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.176415227518705,
|
| 157 |
+
"grad_norm": 3.4313299655914307,
|
| 158 |
+
"learning_rate": 6.1040816969277054e-06,
|
| 159 |
+
"loss": 0.44,
|
| 160 |
+
"step": 100000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.176415227518705,
|
| 164 |
+
"eval_loss": 0.2909528613090515,
|
| 165 |
+
"eval_runtime": 543.2034,
|
| 166 |
+
"eval_samples_per_second": 16.473,
|
| 167 |
+
"eval_steps_per_second": 2.06,
|
| 168 |
+
"eval_wer": 0.27463762217996435,
|
| 169 |
+
"step": 100000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.2940567502705755,
|
| 173 |
+
"grad_norm": 1.1621043682098389,
|
| 174 |
+
"learning_rate": 5.710478245122278e-06,
|
| 175 |
+
"loss": 0.4264,
|
| 176 |
+
"step": 110000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.2940567502705755,
|
| 180 |
+
"eval_loss": 0.2814837098121643,
|
| 181 |
+
"eval_runtime": 547.8907,
|
| 182 |
+
"eval_samples_per_second": 16.332,
|
| 183 |
+
"eval_steps_per_second": 2.042,
|
| 184 |
+
"eval_wer": 0.2683001724023115,
|
| 185 |
+
"step": 110000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.4116982730224459,
|
| 189 |
+
"grad_norm": 2.507716178894043,
|
| 190 |
+
"learning_rate": 5.316914161535675e-06,
|
| 191 |
+
"loss": 0.4189,
|
| 192 |
+
"step": 120000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.4116982730224459,
|
| 196 |
+
"eval_loss": 0.27175650000572205,
|
| 197 |
+
"eval_runtime": 548.1953,
|
| 198 |
+
"eval_samples_per_second": 16.323,
|
| 199 |
+
"eval_steps_per_second": 2.041,
|
| 200 |
+
"eval_wer": 0.26368304611946813,
|
| 201 |
+
"step": 120000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 1.5293397957743164,
|
| 205 |
+
"grad_norm": 2.3356528282165527,
|
| 206 |
+
"learning_rate": 4.923428814386722e-06,
|
| 207 |
+
"loss": 0.4052,
|
| 208 |
+
"step": 130000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.5293397957743164,
|
| 212 |
+
"eval_loss": 0.2673029899597168,
|
| 213 |
+
"eval_runtime": 544.6822,
|
| 214 |
+
"eval_samples_per_second": 16.428,
|
| 215 |
+
"eval_steps_per_second": 2.054,
|
| 216 |
+
"eval_wer": 0.25851467639420195,
|
| 217 |
+
"step": 130000
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 1.646981318526187,
|
| 221 |
+
"grad_norm": 2.476557970046997,
|
| 222 |
+
"learning_rate": 4.529864730800119e-06,
|
| 223 |
+
"loss": 0.4044,
|
| 224 |
+
"step": 140000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.646981318526187,
|
| 228 |
+
"eval_loss": 0.26591917872428894,
|
| 229 |
+
"eval_runtime": 545.1182,
|
| 230 |
+
"eval_samples_per_second": 16.415,
|
| 231 |
+
"eval_steps_per_second": 2.053,
|
| 232 |
+
"eval_wer": 0.2534720937631799,
|
| 233 |
+
"step": 140000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.7646228412780576,
|
| 237 |
+
"grad_norm": 1.8470633029937744,
|
| 238 |
+
"learning_rate": 4.136418751869991e-06,
|
| 239 |
+
"loss": 0.4046,
|
| 240 |
+
"step": 150000
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.7646228412780576,
|
| 244 |
+
"eval_loss": 0.2603091299533844,
|
| 245 |
+
"eval_runtime": 545.5751,
|
| 246 |
+
"eval_samples_per_second": 16.401,
|
| 247 |
+
"eval_steps_per_second": 2.051,
|
| 248 |
+
"eval_wer": 0.24952459877616556,
|
| 249 |
+
"step": 150000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.8822643640299281,
|
| 253 |
+
"grad_norm": 2.6165308952331543,
|
| 254 |
+
"learning_rate": 3.742815300064564e-06,
|
| 255 |
+
"loss": 0.3944,
|
| 256 |
+
"step": 160000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.8822643640299281,
|
| 260 |
+
"eval_loss": 0.2575734555721283,
|
| 261 |
+
"eval_runtime": 545.6155,
|
| 262 |
+
"eval_samples_per_second": 16.4,
|
| 263 |
+
"eval_steps_per_second": 2.051,
|
| 264 |
+
"eval_wer": 0.24606545368445198,
|
| 265 |
+
"step": 160000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 1.9999058867817985,
|
| 269 |
+
"grad_norm": 2.0350422859191895,
|
| 270 |
+
"learning_rate": 3.3492512164779615e-06,
|
| 271 |
+
"loss": 0.3876,
|
| 272 |
+
"step": 170000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.9999058867817985,
|
| 276 |
+
"eval_loss": 0.25538763403892517,
|
| 277 |
+
"eval_runtime": 543.8449,
|
| 278 |
+
"eval_samples_per_second": 16.453,
|
| 279 |
+
"eval_steps_per_second": 2.058,
|
| 280 |
+
"eval_wer": 0.24299106911630866,
|
| 281 |
+
"step": 170000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.117547409533669,
|
| 285 |
+
"grad_norm": 2.234062433242798,
|
| 286 |
+
"learning_rate": 2.955726501110184e-06,
|
| 287 |
+
"loss": 0.3836,
|
| 288 |
+
"step": 180000
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.117547409533669,
|
| 292 |
+
"eval_loss": 0.25172147154808044,
|
| 293 |
+
"eval_runtime": 543.9926,
|
| 294 |
+
"eval_samples_per_second": 16.449,
|
| 295 |
+
"eval_steps_per_second": 2.057,
|
| 296 |
+
"eval_wer": 0.24226224389377649,
|
| 297 |
+
"step": 180000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 2.2351889322855394,
|
| 301 |
+
"grad_norm": 2.7782626152038574,
|
| 302 |
+
"learning_rate": 2.5621624175235817e-06,
|
| 303 |
+
"loss": 0.3767,
|
| 304 |
+
"step": 190000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 2.2351889322855394,
|
| 308 |
+
"eval_loss": 0.2503082752227783,
|
| 309 |
+
"eval_runtime": 546.7703,
|
| 310 |
+
"eval_samples_per_second": 16.365,
|
| 311 |
+
"eval_steps_per_second": 2.047,
|
| 312 |
+
"eval_wer": 0.23942833465286462,
|
| 313 |
+
"step": 190000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 2.35283045503741,
|
| 317 |
+
"grad_norm": 2.365490198135376,
|
| 318 |
+
"learning_rate": 2.1686770703746284e-06,
|
| 319 |
+
"loss": 0.3738,
|
| 320 |
+
"step": 200000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 2.35283045503741,
|
| 324 |
+
"eval_loss": 0.24804162979125977,
|
| 325 |
+
"eval_runtime": 545.3342,
|
| 326 |
+
"eval_samples_per_second": 16.408,
|
| 327 |
+
"eval_steps_per_second": 2.052,
|
| 328 |
+
"eval_wer": 0.23909166919474062,
|
| 329 |
+
"step": 200000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 2.4704719777892805,
|
| 333 |
+
"grad_norm": 3.105099678039551,
|
| 334 |
+
"learning_rate": 1.7751523550068502e-06,
|
| 335 |
+
"loss": 0.3707,
|
| 336 |
+
"step": 210000
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 2.4704719777892805,
|
| 340 |
+
"eval_loss": 0.24506914615631104,
|
| 341 |
+
"eval_runtime": 547.7088,
|
| 342 |
+
"eval_samples_per_second": 16.337,
|
| 343 |
+
"eval_steps_per_second": 2.043,
|
| 344 |
+
"eval_wer": 0.23790409104025928,
|
| 345 |
+
"step": 210000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.588113500541151,
|
| 349 |
+
"grad_norm": 0.9898041486740112,
|
| 350 |
+
"learning_rate": 1.3815882714202481e-06,
|
| 351 |
+
"loss": 0.3649,
|
| 352 |
+
"step": 220000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 2.588113500541151,
|
| 356 |
+
"eval_loss": 0.24370211362838745,
|
| 357 |
+
"eval_runtime": 547.4398,
|
| 358 |
+
"eval_samples_per_second": 16.345,
|
| 359 |
+
"eval_steps_per_second": 2.044,
|
| 360 |
+
"eval_wer": 0.23637614780723498,
|
| 361 |
+
"step": 220000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.7057550232930216,
|
| 365 |
+
"grad_norm": 4.53593635559082,
|
| 366 |
+
"learning_rate": 9.880241878336459e-07,
|
| 367 |
+
"loss": 0.369,
|
| 368 |
+
"step": 230000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.7057550232930216,
|
| 372 |
+
"eval_loss": 0.24421393871307373,
|
| 373 |
+
"eval_runtime": 549.6273,
|
| 374 |
+
"eval_samples_per_second": 16.28,
|
| 375 |
+
"eval_steps_per_second": 2.036,
|
| 376 |
+
"eval_wer": 0.23538094991453876,
|
| 377 |
+
"step": 230000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 2.8233965460448918,
|
| 381 |
+
"grad_norm": 3.995215654373169,
|
| 382 |
+
"learning_rate": 5.945388406846922e-07,
|
| 383 |
+
"loss": 0.3608,
|
| 384 |
+
"step": 240000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.8233965460448918,
|
| 388 |
+
"eval_loss": 0.24260137975215912,
|
| 389 |
+
"eval_runtime": 548.4399,
|
| 390 |
+
"eval_samples_per_second": 16.315,
|
| 391 |
+
"eval_steps_per_second": 2.04,
|
| 392 |
+
"eval_wer": 0.23452633759776248,
|
| 393 |
+
"step": 240000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 2.9410380687967628,
|
| 397 |
+
"grad_norm": 1.9774836301803589,
|
| 398 |
+
"learning_rate": 2.0097475709808989e-07,
|
| 399 |
+
"loss": 0.3644,
|
| 400 |
+
"step": 250000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 2.9410380687967628,
|
| 404 |
+
"eval_loss": 0.24260272085666656,
|
| 405 |
+
"eval_runtime": 548.9964,
|
| 406 |
+
"eval_samples_per_second": 16.299,
|
| 407 |
+
"eval_steps_per_second": 2.038,
|
| 408 |
+
"eval_wer": 0.234019489600367,
|
| 409 |
+
"step": 250000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 3.0,
|
| 413 |
+
"step": 255012,
|
| 414 |
+
"total_flos": 2.7896646019563717e+20,
|
| 415 |
+
"train_loss": 0.5826510110653187,
|
| 416 |
+
"train_runtime": 84804.3792,
|
| 417 |
+
"train_samples_per_second": 6.014,
|
| 418 |
+
"train_steps_per_second": 3.007
|
| 419 |
+
}
|
| 420 |
+
],
|
| 421 |
+
"logging_steps": 10000,
|
| 422 |
+
"max_steps": 255012,
|
| 423 |
+
"num_input_tokens_seen": 0,
|
| 424 |
+
"num_train_epochs": 3,
|
| 425 |
+
"save_steps": 10000,
|
| 426 |
+
"stateful_callbacks": {
|
| 427 |
+
"TrainerControl": {
|
| 428 |
+
"args": {
|
| 429 |
+
"should_epoch_stop": false,
|
| 430 |
+
"should_evaluate": false,
|
| 431 |
+
"should_log": false,
|
| 432 |
+
"should_save": true,
|
| 433 |
+
"should_training_stop": true
|
| 434 |
+
},
|
| 435 |
+
"attributes": {}
|
| 436 |
+
}
|
| 437 |
+
},
|
| 438 |
+
"total_flos": 2.7896646019563717e+20,
|
| 439 |
+
"train_batch_size": 2,
|
| 440 |
+
"trial_name": null,
|
| 441 |
+
"trial_params": null
|
| 442 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:865bee174cb34efab19ff1a39c004b8e7705ebe4b7c71fde74c2b8dc1d0e5280
|
| 3 |
+
size 5176
|