Commit
•
12dff41
1
Parent(s):
132e7fb
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 +322 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.4648422300815582,
|
| 4 |
+
"eval_runtime": 16.9057,
|
| 5 |
+
"eval_samples": 440,
|
| 6 |
+
"eval_samples_per_second": 26.027,
|
| 7 |
+
"eval_steps_per_second": 1.656,
|
| 8 |
+
"eval_wer": 0.5471698113207547,
|
| 9 |
+
"train_loss": 1.7521888146033653,
|
| 10 |
+
"train_runtime": 6573.2529,
|
| 11 |
+
"train_samples": 1220,
|
| 12 |
+
"train_samples_per_second": 18.56,
|
| 13 |
+
"train_steps_per_second": 0.593
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"eval_loss": 0.4648422300815582,
|
| 4 |
+
"eval_runtime": 16.9057,
|
| 5 |
+
"eval_samples": 440,
|
| 6 |
+
"eval_samples_per_second": 26.027,
|
| 7 |
+
"eval_steps_per_second": 1.656,
|
| 8 |
+
"eval_wer": 0.5471698113207547
|
| 9 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1262058993
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1b8b80706ef9a61b6cdc80eb0447f3a10bab0f63c06b9a485688b4fd5ee3a34
|
| 3 |
size 1262058993
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 100.0,
|
| 3 |
+
"train_loss": 1.7521888146033653,
|
| 4 |
+
"train_runtime": 6573.2529,
|
| 5 |
+
"train_samples": 1220,
|
| 6 |
+
"train_samples_per_second": 18.56,
|
| 7 |
+
"train_steps_per_second": 0.593
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,322 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 100.0,
|
| 5 |
+
"global_step": 3900,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 2.56,
|
| 12 |
+
"learning_rate": 1.3439999999999998e-05,
|
| 13 |
+
"loss": 21.7509,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 5.13,
|
| 18 |
+
"learning_rate": 2.7439999999999998e-05,
|
| 19 |
+
"loss": 5.8287,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 7.69,
|
| 24 |
+
"learning_rate": 4.1439999999999996e-05,
|
| 25 |
+
"loss": 3.6855,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 10.26,
|
| 30 |
+
"learning_rate": 5.544e-05,
|
| 31 |
+
"loss": 3.0684,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 12.82,
|
| 36 |
+
"learning_rate": 6.944e-05,
|
| 37 |
+
"loss": 2.9421,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 12.82,
|
| 42 |
+
"eval_loss": 2.889408826828003,
|
| 43 |
+
"eval_runtime": 17.3993,
|
| 44 |
+
"eval_samples_per_second": 25.288,
|
| 45 |
+
"eval_steps_per_second": 1.609,
|
| 46 |
+
"eval_wer": 1.0,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 15.38,
|
| 51 |
+
"learning_rate": 6.802352941176471e-05,
|
| 52 |
+
"loss": 2.847,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 17.95,
|
| 57 |
+
"learning_rate": 6.596470588235294e-05,
|
| 58 |
+
"loss": 2.5861,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 20.51,
|
| 63 |
+
"learning_rate": 6.390588235294118e-05,
|
| 64 |
+
"loss": 1.8427,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 23.08,
|
| 69 |
+
"learning_rate": 6.184705882352941e-05,
|
| 70 |
+
"loss": 1.3659,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 25.64,
|
| 75 |
+
"learning_rate": 5.978823529411764e-05,
|
| 76 |
+
"loss": 1.1872,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 25.64,
|
| 81 |
+
"eval_loss": 0.6688352227210999,
|
| 82 |
+
"eval_runtime": 17.312,
|
| 83 |
+
"eval_samples_per_second": 25.416,
|
| 84 |
+
"eval_steps_per_second": 1.617,
|
| 85 |
+
"eval_wer": 0.7459733087896917,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 28.21,
|
| 90 |
+
"learning_rate": 5.772941176470588e-05,
|
| 91 |
+
"loss": 1.087,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 30.77,
|
| 96 |
+
"learning_rate": 5.5670588235294114e-05,
|
| 97 |
+
"loss": 1.0304,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 33.33,
|
| 102 |
+
"learning_rate": 5.361176470588235e-05,
|
| 103 |
+
"loss": 0.9762,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 35.9,
|
| 108 |
+
"learning_rate": 5.1552941176470585e-05,
|
| 109 |
+
"loss": 0.9043,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 38.46,
|
| 114 |
+
"learning_rate": 4.949411764705882e-05,
|
| 115 |
+
"loss": 0.8894,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 38.46,
|
| 120 |
+
"eval_loss": 0.4868045449256897,
|
| 121 |
+
"eval_runtime": 16.7955,
|
| 122 |
+
"eval_samples_per_second": 26.197,
|
| 123 |
+
"eval_steps_per_second": 1.667,
|
| 124 |
+
"eval_wer": 0.6516336861481823,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 41.03,
|
| 129 |
+
"learning_rate": 4.7435294117647055e-05,
|
| 130 |
+
"loss": 0.861,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 43.59,
|
| 135 |
+
"learning_rate": 4.5376470588235284e-05,
|
| 136 |
+
"loss": 0.8411,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 46.15,
|
| 141 |
+
"learning_rate": 4.331764705882352e-05,
|
| 142 |
+
"loss": 0.8349,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 48.72,
|
| 147 |
+
"learning_rate": 4.1258823529411755e-05,
|
| 148 |
+
"loss": 0.7912,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 51.28,
|
| 153 |
+
"learning_rate": 3.92e-05,
|
| 154 |
+
"loss": 0.769,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 51.28,
|
| 159 |
+
"eval_loss": 0.495985209941864,
|
| 160 |
+
"eval_runtime": 17.6095,
|
| 161 |
+
"eval_samples_per_second": 24.987,
|
| 162 |
+
"eval_steps_per_second": 1.59,
|
| 163 |
+
"eval_wer": 0.650713299585826,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 53.85,
|
| 168 |
+
"learning_rate": 3.714117647058823e-05,
|
| 169 |
+
"loss": 0.7639,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 56.41,
|
| 174 |
+
"learning_rate": 3.508235294117647e-05,
|
| 175 |
+
"loss": 0.7421,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 58.97,
|
| 180 |
+
"learning_rate": 3.3023529411764704e-05,
|
| 181 |
+
"loss": 0.728,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 61.54,
|
| 186 |
+
"learning_rate": 3.096470588235294e-05,
|
| 187 |
+
"loss": 0.7105,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 64.1,
|
| 192 |
+
"learning_rate": 2.8905882352941175e-05,
|
| 193 |
+
"loss": 0.6936,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 64.1,
|
| 198 |
+
"eval_loss": 0.47812411189079285,
|
| 199 |
+
"eval_runtime": 17.7238,
|
| 200 |
+
"eval_samples_per_second": 24.825,
|
| 201 |
+
"eval_steps_per_second": 1.58,
|
| 202 |
+
"eval_wer": 0.5384261389783709,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 66.67,
|
| 207 |
+
"learning_rate": 2.684705882352941e-05,
|
| 208 |
+
"loss": 0.6813,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 69.23,
|
| 213 |
+
"learning_rate": 2.4788235294117646e-05,
|
| 214 |
+
"loss": 0.6514,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 71.79,
|
| 219 |
+
"learning_rate": 2.272941176470588e-05,
|
| 220 |
+
"loss": 0.6539,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 74.36,
|
| 225 |
+
"learning_rate": 2.0670588235294117e-05,
|
| 226 |
+
"loss": 0.6423,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 76.92,
|
| 231 |
+
"learning_rate": 1.8611764705882352e-05,
|
| 232 |
+
"loss": 0.624,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 76.92,
|
| 237 |
+
"eval_loss": 0.4642501175403595,
|
| 238 |
+
"eval_runtime": 16.6285,
|
| 239 |
+
"eval_samples_per_second": 26.461,
|
| 240 |
+
"eval_steps_per_second": 1.684,
|
| 241 |
+
"eval_wer": 0.5430280717901519,
|
| 242 |
+
"step": 3000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 79.49,
|
| 246 |
+
"learning_rate": 1.6552941176470588e-05,
|
| 247 |
+
"loss": 0.624,
|
| 248 |
+
"step": 3100
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 82.05,
|
| 252 |
+
"learning_rate": 1.4494117647058823e-05,
|
| 253 |
+
"loss": 0.6071,
|
| 254 |
+
"step": 3200
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 84.62,
|
| 258 |
+
"learning_rate": 1.2435294117647057e-05,
|
| 259 |
+
"loss": 0.6175,
|
| 260 |
+
"step": 3300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 87.18,
|
| 264 |
+
"learning_rate": 1.0376470588235292e-05,
|
| 265 |
+
"loss": 0.6023,
|
| 266 |
+
"step": 3400
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 89.74,
|
| 270 |
+
"learning_rate": 8.317647058823528e-06,
|
| 271 |
+
"loss": 0.5966,
|
| 272 |
+
"step": 3500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 89.74,
|
| 276 |
+
"eval_loss": 0.4529646337032318,
|
| 277 |
+
"eval_runtime": 16.5674,
|
| 278 |
+
"eval_samples_per_second": 26.558,
|
| 279 |
+
"eval_steps_per_second": 1.69,
|
| 280 |
+
"eval_wer": 0.5591348366313852,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 92.31,
|
| 285 |
+
"learning_rate": 6.258823529411764e-06,
|
| 286 |
+
"loss": 0.5868,
|
| 287 |
+
"step": 3600
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 94.87,
|
| 291 |
+
"learning_rate": 4.2e-06,
|
| 292 |
+
"loss": 0.5817,
|
| 293 |
+
"step": 3700
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 97.44,
|
| 297 |
+
"learning_rate": 2.141176470588235e-06,
|
| 298 |
+
"loss": 0.5687,
|
| 299 |
+
"step": 3800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 100.0,
|
| 303 |
+
"learning_rate": 8.235294117647057e-08,
|
| 304 |
+
"loss": 0.5709,
|
| 305 |
+
"step": 3900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 100.0,
|
| 309 |
+
"step": 3900,
|
| 310 |
+
"total_flos": 1.7166559536300583e+19,
|
| 311 |
+
"train_loss": 1.7521888146033653,
|
| 312 |
+
"train_runtime": 6573.2529,
|
| 313 |
+
"train_samples_per_second": 18.56,
|
| 314 |
+
"train_steps_per_second": 0.593
|
| 315 |
+
}
|
| 316 |
+
],
|
| 317 |
+
"max_steps": 3900,
|
| 318 |
+
"num_train_epochs": 100,
|
| 319 |
+
"total_flos": 1.7166559536300583e+19,
|
| 320 |
+
"trial_name": null,
|
| 321 |
+
"trial_params": null
|
| 322 |
+
}
|