Commit
•
5abce05
1
Parent(s):
dd5117d
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 13.06,
|
| 3 |
+
"eval_loss": 0.11871337890625,
|
| 4 |
+
"eval_runtime": 3666.7152,
|
| 5 |
+
"eval_samples_per_second": 0.682,
|
| 6 |
+
"eval_steps_per_second": 0.043,
|
| 7 |
+
"eval_wer": 31.85479597244303,
|
| 8 |
+
"train_loss": 0.04660231500864029,
|
| 9 |
+
"train_runtime": 35334.3457,
|
| 10 |
+
"train_samples_per_second": 1.811,
|
| 11 |
+
"train_steps_per_second": 0.028
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 13.06,
|
| 3 |
+
"eval_loss": 0.11871337890625,
|
| 4 |
+
"eval_runtime": 3666.7152,
|
| 5 |
+
"eval_samples_per_second": 0.682,
|
| 6 |
+
"eval_steps_per_second": 0.043,
|
| 7 |
+
"eval_wer": 31.85479597244303
|
| 8 |
+
}
|
runs/Dec15_09-02-52_129-146-105-103/events.out.tfevents.1671134227.129-146-105-103.2422739.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f50ce93d15f3a244e999ce1f0e72afcf3bccada86d85d092fb31fc9e487f2db0
|
| 3 |
+
size 358
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 13.06,
|
| 3 |
+
"train_loss": 0.04660231500864029,
|
| 4 |
+
"train_runtime": 35334.3457,
|
| 5 |
+
"train_samples_per_second": 1.811,
|
| 6 |
+
"train_steps_per_second": 0.028
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,310 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 31.85479597244303,
|
| 3 |
+
"best_model_checkpoint": "./checkpoint-1000",
|
| 4 |
+
"epoch": 13.064,
|
| 5 |
+
"global_step": 1000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 2.2e-06,
|
| 13 |
+
"loss": 0.9495,
|
| 14 |
+
"step": 25
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 4.7e-06,
|
| 19 |
+
"loss": 0.2746,
|
| 20 |
+
"step": 50
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.0,
|
| 24 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 25 |
+
"loss": 0.15,
|
| 26 |
+
"step": 75
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.03,
|
| 30 |
+
"learning_rate": 9.7e-06,
|
| 31 |
+
"loss": 0.0876,
|
| 32 |
+
"step": 100
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.05,
|
| 36 |
+
"learning_rate": 9.755555555555556e-06,
|
| 37 |
+
"loss": 0.0795,
|
| 38 |
+
"step": 125
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 2.01,
|
| 42 |
+
"learning_rate": 9.466666666666667e-06,
|
| 43 |
+
"loss": 0.0743,
|
| 44 |
+
"step": 150
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 2.03,
|
| 48 |
+
"learning_rate": 9.188888888888889e-06,
|
| 49 |
+
"loss": 0.0431,
|
| 50 |
+
"step": 175
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.06,
|
| 54 |
+
"learning_rate": 8.91111111111111e-06,
|
| 55 |
+
"loss": 0.0413,
|
| 56 |
+
"step": 200
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 2.06,
|
| 60 |
+
"eval_loss": 0.0711669921875,
|
| 61 |
+
"eval_runtime": 3657.1623,
|
| 62 |
+
"eval_samples_per_second": 0.683,
|
| 63 |
+
"eval_steps_per_second": 0.043,
|
| 64 |
+
"eval_wer": 36.629570747217805,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 3.01,
|
| 69 |
+
"learning_rate": 8.633333333333334e-06,
|
| 70 |
+
"loss": 0.037,
|
| 71 |
+
"step": 225
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 3.03,
|
| 75 |
+
"learning_rate": 8.355555555555556e-06,
|
| 76 |
+
"loss": 0.0223,
|
| 77 |
+
"step": 250
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 3.06,
|
| 81 |
+
"learning_rate": 8.077777777777778e-06,
|
| 82 |
+
"loss": 0.0204,
|
| 83 |
+
"step": 275
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 4.01,
|
| 87 |
+
"learning_rate": 7.788888888888889e-06,
|
| 88 |
+
"loss": 0.0189,
|
| 89 |
+
"step": 300
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 4.04,
|
| 93 |
+
"learning_rate": 7.511111111111111e-06,
|
| 94 |
+
"loss": 0.0121,
|
| 95 |
+
"step": 325
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 4.06,
|
| 99 |
+
"learning_rate": 7.233333333333334e-06,
|
| 100 |
+
"loss": 0.0116,
|
| 101 |
+
"step": 350
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 5.01,
|
| 105 |
+
"learning_rate": 6.955555555555557e-06,
|
| 106 |
+
"loss": 0.009,
|
| 107 |
+
"step": 375
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 5.04,
|
| 111 |
+
"learning_rate": 6.677777777777779e-06,
|
| 112 |
+
"loss": 0.006,
|
| 113 |
+
"step": 400
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 5.04,
|
| 117 |
+
"eval_loss": 0.08990478515625,
|
| 118 |
+
"eval_runtime": 3646.6012,
|
| 119 |
+
"eval_samples_per_second": 0.685,
|
| 120 |
+
"eval_steps_per_second": 0.043,
|
| 121 |
+
"eval_wer": 33.74668786433492,
|
| 122 |
+
"step": 400
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 5.07,
|
| 126 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 127 |
+
"loss": 0.0063,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 6.02,
|
| 132 |
+
"learning_rate": 6.111111111111112e-06,
|
| 133 |
+
"loss": 0.0043,
|
| 134 |
+
"step": 450
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.04,
|
| 138 |
+
"learning_rate": 5.833333333333334e-06,
|
| 139 |
+
"loss": 0.0031,
|
| 140 |
+
"step": 475
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 6.07,
|
| 144 |
+
"learning_rate": 5.555555555555557e-06,
|
| 145 |
+
"loss": 0.0031,
|
| 146 |
+
"step": 500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 7.02,
|
| 150 |
+
"learning_rate": 5.2777777777777785e-06,
|
| 151 |
+
"loss": 0.0021,
|
| 152 |
+
"step": 525
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 7.05,
|
| 156 |
+
"learning_rate": 5e-06,
|
| 157 |
+
"loss": 0.0016,
|
| 158 |
+
"step": 550
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 7.07,
|
| 162 |
+
"learning_rate": 4.722222222222222e-06,
|
| 163 |
+
"loss": 0.0015,
|
| 164 |
+
"step": 575
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 8.02,
|
| 168 |
+
"learning_rate": 4.433333333333334e-06,
|
| 169 |
+
"loss": 0.0008,
|
| 170 |
+
"step": 600
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 8.02,
|
| 174 |
+
"eval_loss": 0.10394287109375,
|
| 175 |
+
"eval_runtime": 3647.236,
|
| 176 |
+
"eval_samples_per_second": 0.685,
|
| 177 |
+
"eval_steps_per_second": 0.043,
|
| 178 |
+
"eval_wer": 32.231054583995764,
|
| 179 |
+
"step": 600
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 8.05,
|
| 183 |
+
"learning_rate": 4.155555555555556e-06,
|
| 184 |
+
"loss": 0.0009,
|
| 185 |
+
"step": 625
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 9.0,
|
| 189 |
+
"learning_rate": 3.877777777777778e-06,
|
| 190 |
+
"loss": 0.0007,
|
| 191 |
+
"step": 650
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 9.03,
|
| 195 |
+
"learning_rate": 3.6000000000000003e-06,
|
| 196 |
+
"loss": 0.0004,
|
| 197 |
+
"step": 675
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 9.05,
|
| 201 |
+
"learning_rate": 3.322222222222222e-06,
|
| 202 |
+
"loss": 0.0004,
|
| 203 |
+
"step": 700
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 10.01,
|
| 207 |
+
"learning_rate": 3.0333333333333337e-06,
|
| 208 |
+
"loss": 0.0004,
|
| 209 |
+
"step": 725
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 10.03,
|
| 213 |
+
"learning_rate": 2.755555555555556e-06,
|
| 214 |
+
"loss": 0.0003,
|
| 215 |
+
"step": 750
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 10.05,
|
| 219 |
+
"learning_rate": 2.4777777777777782e-06,
|
| 220 |
+
"loss": 0.0003,
|
| 221 |
+
"step": 775
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 11.01,
|
| 225 |
+
"learning_rate": 2.2e-06,
|
| 226 |
+
"loss": 0.0002,
|
| 227 |
+
"step": 800
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 11.01,
|
| 231 |
+
"eval_loss": 0.1141357421875,
|
| 232 |
+
"eval_runtime": 3640.5182,
|
| 233 |
+
"eval_samples_per_second": 0.686,
|
| 234 |
+
"eval_steps_per_second": 0.043,
|
| 235 |
+
"eval_wer": 31.928987811340754,
|
| 236 |
+
"step": 800
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 11.03,
|
| 240 |
+
"learning_rate": 1.9222222222222224e-06,
|
| 241 |
+
"loss": 0.0001,
|
| 242 |
+
"step": 825
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 11.06,
|
| 246 |
+
"learning_rate": 1.6444444444444447e-06,
|
| 247 |
+
"loss": 0.0001,
|
| 248 |
+
"step": 850
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 12.01,
|
| 252 |
+
"learning_rate": 1.3555555555555558e-06,
|
| 253 |
+
"loss": 0.0001,
|
| 254 |
+
"step": 875
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 12.04,
|
| 258 |
+
"learning_rate": 1.0777777777777779e-06,
|
| 259 |
+
"loss": 0.0001,
|
| 260 |
+
"step": 900
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 12.06,
|
| 264 |
+
"learning_rate": 8.000000000000001e-07,
|
| 265 |
+
"loss": 0.0001,
|
| 266 |
+
"step": 925
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 13.01,
|
| 270 |
+
"learning_rate": 5.222222222222223e-07,
|
| 271 |
+
"loss": 0.0001,
|
| 272 |
+
"step": 950
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 13.04,
|
| 276 |
+
"learning_rate": 2.444444444444445e-07,
|
| 277 |
+
"loss": 0.0001,
|
| 278 |
+
"step": 975
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 13.06,
|
| 282 |
+
"learning_rate": 0.0,
|
| 283 |
+
"loss": 0.0001,
|
| 284 |
+
"step": 1000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 13.06,
|
| 288 |
+
"eval_loss": 0.11871337890625,
|
| 289 |
+
"eval_runtime": 3861.227,
|
| 290 |
+
"eval_samples_per_second": 0.647,
|
| 291 |
+
"eval_steps_per_second": 0.041,
|
| 292 |
+
"eval_wer": 31.85479597244303,
|
| 293 |
+
"step": 1000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 13.06,
|
| 297 |
+
"step": 1000,
|
| 298 |
+
"total_flos": 1.3615895701052457e+20,
|
| 299 |
+
"train_loss": 0.04660231500864029,
|
| 300 |
+
"train_runtime": 35334.3457,
|
| 301 |
+
"train_samples_per_second": 1.811,
|
| 302 |
+
"train_steps_per_second": 0.028
|
| 303 |
+
}
|
| 304 |
+
],
|
| 305 |
+
"max_steps": 1000,
|
| 306 |
+
"num_train_epochs": 9223372036854775807,
|
| 307 |
+
"total_flos": 1.3615895701052457e+20,
|
| 308 |
+
"trial_name": null,
|
| 309 |
+
"trial_params": null
|
| 310 |
+
}
|