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 +283 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.8003901243209839,
|
| 4 |
+
"eval_runtime": 244.1765,
|
| 5 |
+
"eval_samples": 4620,
|
| 6 |
+
"eval_samples_per_second": 18.921,
|
| 7 |
+
"eval_steps_per_second": 4.73,
|
| 8 |
+
"eval_wer": 0.7138530257602643,
|
| 9 |
+
"train_loss": 2.1586562755496,
|
| 10 |
+
"train_runtime": 12185.0109,
|
| 11 |
+
"train_samples": 11030,
|
| 12 |
+
"train_samples_per_second": 9.052,
|
| 13 |
+
"train_steps_per_second": 0.282
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.8003901243209839,
|
| 4 |
+
"eval_runtime": 244.1765,
|
| 5 |
+
"eval_samples": 4620,
|
| 6 |
+
"eval_samples_per_second": 18.921,
|
| 7 |
+
"eval_steps_per_second": 4.73,
|
| 8 |
+
"eval_wer": 0.7138530257602643
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262075377
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fab04453ef31717d5c79f49a9ab4367b942ff7af5783e8728af4ad4886acf678
|
| 3 |
size 1262075377
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 2.1586562755496,
|
| 4 |
+
"train_runtime": 12185.0109,
|
| 5 |
+
"train_samples": 11030,
|
| 6 |
+
"train_samples_per_second": 9.052,
|
| 7 |
+
"train_steps_per_second": 0.282
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,283 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.997824510514866,
|
| 5 |
+
"global_step": 3440,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.29,
|
| 12 |
+
"learning_rate": 5e-05,
|
| 13 |
+
"loss": 7.294,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.58,
|
| 18 |
+
"learning_rate": 0.0001,
|
| 19 |
+
"loss": 3.2282,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.87,
|
| 24 |
+
"learning_rate": 0.00015,
|
| 25 |
+
"loss": 3.0143,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.16,
|
| 30 |
+
"learning_rate": 0.0002,
|
| 31 |
+
"loss": 2.9501,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.45,
|
| 36 |
+
"learning_rate": 0.00025,
|
| 37 |
+
"loss": 2.6683,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.45,
|
| 42 |
+
"eval_loss": 1.7697819471359253,
|
| 43 |
+
"eval_runtime": 250.7581,
|
| 44 |
+
"eval_samples_per_second": 18.424,
|
| 45 |
+
"eval_steps_per_second": 4.606,
|
| 46 |
+
"eval_wer": 1.0040690203756961,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.74,
|
| 51 |
+
"learning_rate": 0.0003,
|
| 52 |
+
"loss": 2.1972,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.03,
|
| 57 |
+
"learning_rate": 0.00035,
|
| 58 |
+
"loss": 2.0775,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.32,
|
| 63 |
+
"learning_rate": 0.0004,
|
| 64 |
+
"loss": 1.9871,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 2.61,
|
| 69 |
+
"learning_rate": 0.00045000000000000004,
|
| 70 |
+
"loss": 1.9768,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 2.91,
|
| 75 |
+
"learning_rate": 0.0005,
|
| 76 |
+
"loss": 1.9548,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2.91,
|
| 81 |
+
"eval_loss": 1.0889918804168701,
|
| 82 |
+
"eval_runtime": 243.6531,
|
| 83 |
+
"eval_samples_per_second": 18.961,
|
| 84 |
+
"eval_steps_per_second": 4.74,
|
| 85 |
+
"eval_wer": 0.8601847885945053,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.2,
|
| 90 |
+
"learning_rate": 0.00055,
|
| 91 |
+
"loss": 1.9588,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.49,
|
| 96 |
+
"learning_rate": 0.0006,
|
| 97 |
+
"loss": 1.9734,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 3.78,
|
| 102 |
+
"learning_rate": 0.0006495,
|
| 103 |
+
"loss": 1.9725,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.07,
|
| 108 |
+
"learning_rate": 0.0006995,
|
| 109 |
+
"loss": 1.9717,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.36,
|
| 114 |
+
"learning_rate": 0.0007495000000000001,
|
| 115 |
+
"loss": 1.9568,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 4.36,
|
| 120 |
+
"eval_loss": 1.0877875089645386,
|
| 121 |
+
"eval_runtime": 239.5705,
|
| 122 |
+
"eval_samples_per_second": 19.285,
|
| 123 |
+
"eval_steps_per_second": 4.821,
|
| 124 |
+
"eval_wer": 0.868016887964266,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.65,
|
| 129 |
+
"learning_rate": 0.0007995,
|
| 130 |
+
"loss": 1.9528,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 4.94,
|
| 135 |
+
"learning_rate": 0.0008495000000000001,
|
| 136 |
+
"loss": 1.9862,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 5.23,
|
| 141 |
+
"learning_rate": 0.0008995,
|
| 142 |
+
"loss": 1.9592,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 5.52,
|
| 147 |
+
"learning_rate": 0.0009495,
|
| 148 |
+
"loss": 1.938,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 5.81,
|
| 153 |
+
"learning_rate": 0.0009995000000000002,
|
| 154 |
+
"loss": 1.9497,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 5.81,
|
| 159 |
+
"eval_loss": 1.1500531435012817,
|
| 160 |
+
"eval_runtime": 242.7035,
|
| 161 |
+
"eval_samples_per_second": 19.036,
|
| 162 |
+
"eval_steps_per_second": 4.759,
|
| 163 |
+
"eval_wer": 0.8837728691182769,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 6.1,
|
| 168 |
+
"learning_rate": 0.00093125,
|
| 169 |
+
"loss": 1.9326,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 6.39,
|
| 174 |
+
"learning_rate": 0.0008618055555555557,
|
| 175 |
+
"loss": 1.8927,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.68,
|
| 180 |
+
"learning_rate": 0.0007923611111111111,
|
| 181 |
+
"loss": 1.8929,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.97,
|
| 186 |
+
"learning_rate": 0.0007229166666666666,
|
| 187 |
+
"loss": 1.8684,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 7.27,
|
| 192 |
+
"learning_rate": 0.0006534722222222223,
|
| 193 |
+
"loss": 1.8453,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 7.27,
|
| 198 |
+
"eval_loss": 1.0452075004577637,
|
| 199 |
+
"eval_runtime": 247.1245,
|
| 200 |
+
"eval_samples_per_second": 18.695,
|
| 201 |
+
"eval_steps_per_second": 4.674,
|
| 202 |
+
"eval_wer": 0.8417977115584654,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 7.56,
|
| 207 |
+
"learning_rate": 0.0005840277777777778,
|
| 208 |
+
"loss": 1.8081,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 7.85,
|
| 213 |
+
"learning_rate": 0.0005145833333333333,
|
| 214 |
+
"loss": 1.78,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 8.14,
|
| 219 |
+
"learning_rate": 0.00044513888888888885,
|
| 220 |
+
"loss": 1.7618,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 8.43,
|
| 225 |
+
"learning_rate": 0.00037569444444444445,
|
| 226 |
+
"loss": 1.7155,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 8.72,
|
| 231 |
+
"learning_rate": 0.00030625000000000004,
|
| 232 |
+
"loss": 1.6952,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 8.72,
|
| 237 |
+
"eval_loss": 0.9152895212173462,
|
| 238 |
+
"eval_runtime": 242.409,
|
| 239 |
+
"eval_samples_per_second": 19.059,
|
| 240 |
+
"eval_steps_per_second": 4.765,
|
| 241 |
+
"eval_wer": 0.7822615186930184,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 9.01,
|
| 246 |
+
"learning_rate": 0.00023680555555555556,
|
| 247 |
+
"loss": 1.6874,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 9.3,
|
| 252 |
+
"learning_rate": 0.0001673611111111111,
|
| 253 |
+
"loss": 1.6143,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 9.59,
|
| 258 |
+
"learning_rate": 9.791666666666667e-05,
|
| 259 |
+
"loss": 1.5948,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 9.88,
|
| 264 |
+
"learning_rate": 2.9166666666666666e-05,
|
| 265 |
+
"loss": 1.5723,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 10.0,
|
| 270 |
+
"step": 3440,
|
| 271 |
+
"total_flos": 1.01153307245466e+19,
|
| 272 |
+
"train_loss": 2.1586562755496,
|
| 273 |
+
"train_runtime": 12185.0109,
|
| 274 |
+
"train_samples_per_second": 9.052,
|
| 275 |
+
"train_steps_per_second": 0.282
|
| 276 |
+
}
|
| 277 |
+
],
|
| 278 |
+
"max_steps": 3440,
|
| 279 |
+
"num_train_epochs": 10,
|
| 280 |
+
"total_flos": 1.01153307245466e+19,
|
| 281 |
+
"trial_name": null,
|
| 282 |
+
"trial_params": null
|
| 283 |
+
}
|