Commit
•
8dfc6af
1
Parent(s):
a700171
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.99,
|
| 3 |
+
"eval_loss": 0.4357174038887024,
|
| 4 |
+
"eval_runtime": 110.7387,
|
| 5 |
+
"eval_samples": 3876,
|
| 6 |
+
"eval_samples_per_second": 35.001,
|
| 7 |
+
"eval_steps_per_second": 4.38,
|
| 8 |
+
"eval_wer": 0.3262315072590479,
|
| 9 |
+
"train_loss": 1.0930316641311015,
|
| 10 |
+
"train_runtime": 6038.6358,
|
| 11 |
+
"train_samples": 15504,
|
| 12 |
+
"train_samples_per_second": 12.837,
|
| 13 |
+
"train_steps_per_second": 0.401
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.99,
|
| 3 |
+
"eval_loss": 0.4357174038887024,
|
| 4 |
+
"eval_runtime": 110.7387,
|
| 5 |
+
"eval_samples": 3876,
|
| 6 |
+
"eval_samples_per_second": 35.001,
|
| 7 |
+
"eval_steps_per_second": 4.38,
|
| 8 |
+
"eval_wer": 0.3262315072590479
|
| 9 |
+
}
|
runs/May15_11-49-42_d5de9a2368d8/events.out.tfevents.1684154777.d5de9a2368d8.1059.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b2d2e70b2617602075bb7005cd7434082917b959662facae4908ea3357f2196c
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.99,
|
| 3 |
+
"train_loss": 1.0930316641311015,
|
| 4 |
+
"train_runtime": 6038.6358,
|
| 5 |
+
"train_samples": 15504,
|
| 6 |
+
"train_samples_per_second": 12.837,
|
| 7 |
+
"train_steps_per_second": 0.401
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.994840041279669,
|
| 5 |
+
"global_step": 2420,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.21,
|
| 12 |
+
"eval_loss": 3.4135334491729736,
|
| 13 |
+
"eval_runtime": 150.5586,
|
| 14 |
+
"eval_samples_per_second": 25.744,
|
| 15 |
+
"eval_steps_per_second": 3.221,
|
| 16 |
+
"eval_wer": 1.0,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.41,
|
| 21 |
+
"eval_loss": 2.9520680904388428,
|
| 22 |
+
"eval_runtime": 111.0015,
|
| 23 |
+
"eval_samples_per_second": 34.918,
|
| 24 |
+
"eval_steps_per_second": 4.369,
|
| 25 |
+
"eval_wer": 1.0,
|
| 26 |
+
"step": 200
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.62,
|
| 30 |
+
"eval_loss": 2.333911418914795,
|
| 31 |
+
"eval_runtime": 110.7137,
|
| 32 |
+
"eval_samples_per_second": 35.009,
|
| 33 |
+
"eval_steps_per_second": 4.381,
|
| 34 |
+
"eval_wer": 0.9364516945447853,
|
| 35 |
+
"step": 300
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.83,
|
| 39 |
+
"eval_loss": 1.2433439493179321,
|
| 40 |
+
"eval_runtime": 111.6882,
|
| 41 |
+
"eval_samples_per_second": 34.704,
|
| 42 |
+
"eval_steps_per_second": 4.342,
|
| 43 |
+
"eval_wer": 0.8259439063111153,
|
| 44 |
+
"step": 400
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.03,
|
| 48 |
+
"learning_rate": 0.00029759999999999997,
|
| 49 |
+
"loss": 3.1912,
|
| 50 |
+
"step": 500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.03,
|
| 54 |
+
"eval_loss": 0.8614437580108643,
|
| 55 |
+
"eval_runtime": 110.721,
|
| 56 |
+
"eval_samples_per_second": 35.007,
|
| 57 |
+
"eval_steps_per_second": 4.38,
|
| 58 |
+
"eval_wer": 0.6385201205623173,
|
| 59 |
+
"step": 500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.24,
|
| 63 |
+
"eval_loss": 0.7556686401367188,
|
| 64 |
+
"eval_runtime": 111.0146,
|
| 65 |
+
"eval_samples_per_second": 34.914,
|
| 66 |
+
"eval_steps_per_second": 4.369,
|
| 67 |
+
"eval_wer": 0.5611899776821664,
|
| 68 |
+
"step": 600
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.44,
|
| 72 |
+
"eval_loss": 0.6781343221664429,
|
| 73 |
+
"eval_runtime": 110.9557,
|
| 74 |
+
"eval_samples_per_second": 34.933,
|
| 75 |
+
"eval_steps_per_second": 4.371,
|
| 76 |
+
"eval_wer": 0.5194533281181695,
|
| 77 |
+
"step": 700
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.65,
|
| 81 |
+
"eval_loss": 0.6363114714622498,
|
| 82 |
+
"eval_runtime": 111.5716,
|
| 83 |
+
"eval_samples_per_second": 34.74,
|
| 84 |
+
"eval_steps_per_second": 4.347,
|
| 85 |
+
"eval_wer": 0.4878862480730736,
|
| 86 |
+
"step": 800
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.86,
|
| 90 |
+
"eval_loss": 0.5959396362304688,
|
| 91 |
+
"eval_runtime": 111.2645,
|
| 92 |
+
"eval_samples_per_second": 34.836,
|
| 93 |
+
"eval_steps_per_second": 4.359,
|
| 94 |
+
"eval_wer": 0.4559050226629547,
|
| 95 |
+
"step": 900
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.06,
|
| 99 |
+
"learning_rate": 0.00022281249999999997,
|
| 100 |
+
"loss": 0.8237,
|
| 101 |
+
"step": 1000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.06,
|
| 105 |
+
"eval_loss": 0.5430020093917847,
|
| 106 |
+
"eval_runtime": 111.36,
|
| 107 |
+
"eval_samples_per_second": 34.806,
|
| 108 |
+
"eval_steps_per_second": 4.355,
|
| 109 |
+
"eval_wer": 0.42597151600211675,
|
| 110 |
+
"step": 1000
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.27,
|
| 114 |
+
"eval_loss": 0.5292537212371826,
|
| 115 |
+
"eval_runtime": 111.5324,
|
| 116 |
+
"eval_samples_per_second": 34.752,
|
| 117 |
+
"eval_steps_per_second": 4.349,
|
| 118 |
+
"eval_wer": 0.4097738306145457,
|
| 119 |
+
"step": 1100
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 2.48,
|
| 123 |
+
"eval_loss": 0.5140984654426575,
|
| 124 |
+
"eval_runtime": 111.6965,
|
| 125 |
+
"eval_samples_per_second": 34.701,
|
| 126 |
+
"eval_steps_per_second": 4.342,
|
| 127 |
+
"eval_wer": 0.40558636081264526,
|
| 128 |
+
"step": 1200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.68,
|
| 132 |
+
"eval_loss": 0.4878837764263153,
|
| 133 |
+
"eval_runtime": 110.9007,
|
| 134 |
+
"eval_samples_per_second": 34.95,
|
| 135 |
+
"eval_steps_per_second": 4.373,
|
| 136 |
+
"eval_wer": 0.3946805328670363,
|
| 137 |
+
"step": 1300
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.89,
|
| 141 |
+
"eval_loss": 0.46965792775154114,
|
| 142 |
+
"eval_runtime": 112.2618,
|
| 143 |
+
"eval_samples_per_second": 34.526,
|
| 144 |
+
"eval_steps_per_second": 4.32,
|
| 145 |
+
"eval_wer": 0.37882796861698453,
|
| 146 |
+
"step": 1400
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.1,
|
| 150 |
+
"learning_rate": 0.000145,
|
| 151 |
+
"loss": 0.5625,
|
| 152 |
+
"step": 1500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 3.1,
|
| 156 |
+
"eval_loss": 0.4748215675354004,
|
| 157 |
+
"eval_runtime": 111.3137,
|
| 158 |
+
"eval_samples_per_second": 34.821,
|
| 159 |
+
"eval_steps_per_second": 4.357,
|
| 160 |
+
"eval_wer": 0.37799967788693833,
|
| 161 |
+
"step": 1500
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 3.3,
|
| 165 |
+
"eval_loss": 0.48358285427093506,
|
| 166 |
+
"eval_runtime": 111.983,
|
| 167 |
+
"eval_samples_per_second": 34.612,
|
| 168 |
+
"eval_steps_per_second": 4.331,
|
| 169 |
+
"eval_wer": 0.368382302188068,
|
| 170 |
+
"step": 1600
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.51,
|
| 174 |
+
"eval_loss": 0.47962862253189087,
|
| 175 |
+
"eval_runtime": 112.0713,
|
| 176 |
+
"eval_samples_per_second": 34.585,
|
| 177 |
+
"eval_steps_per_second": 4.328,
|
| 178 |
+
"eval_wer": 0.36251524285024045,
|
| 179 |
+
"step": 1700
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 3.72,
|
| 183 |
+
"eval_loss": 0.4582051932811737,
|
| 184 |
+
"eval_runtime": 112.8803,
|
| 185 |
+
"eval_samples_per_second": 34.337,
|
| 186 |
+
"eval_steps_per_second": 4.297,
|
| 187 |
+
"eval_wer": 0.35147136644962385,
|
| 188 |
+
"step": 1800
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 3.92,
|
| 192 |
+
"eval_loss": 0.43948230147361755,
|
| 193 |
+
"eval_runtime": 112.0141,
|
| 194 |
+
"eval_samples_per_second": 34.603,
|
| 195 |
+
"eval_steps_per_second": 4.33,
|
| 196 |
+
"eval_wer": 0.3437406529691922,
|
| 197 |
+
"step": 1900
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 4.13,
|
| 201 |
+
"learning_rate": 6.71875e-05,
|
| 202 |
+
"loss": 0.4267,
|
| 203 |
+
"step": 2000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 4.13,
|
| 207 |
+
"eval_loss": 0.44096100330352783,
|
| 208 |
+
"eval_runtime": 111.9224,
|
| 209 |
+
"eval_samples_per_second": 34.631,
|
| 210 |
+
"eval_steps_per_second": 4.333,
|
| 211 |
+
"eval_wer": 0.3420150472815958,
|
| 212 |
+
"step": 2000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 4.33,
|
| 216 |
+
"eval_loss": 0.4466901123523712,
|
| 217 |
+
"eval_runtime": 112.4959,
|
| 218 |
+
"eval_samples_per_second": 34.455,
|
| 219 |
+
"eval_steps_per_second": 4.311,
|
| 220 |
+
"eval_wer": 0.3382187147688839,
|
| 221 |
+
"step": 2100
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 4.54,
|
| 225 |
+
"eval_loss": 0.4398203492164612,
|
| 226 |
+
"eval_runtime": 113.1035,
|
| 227 |
+
"eval_samples_per_second": 34.269,
|
| 228 |
+
"eval_steps_per_second": 4.288,
|
| 229 |
+
"eval_wer": 0.33292685732692173,
|
| 230 |
+
"step": 2200
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 4.75,
|
| 234 |
+
"eval_loss": 0.4382670521736145,
|
| 235 |
+
"eval_runtime": 112.3465,
|
| 236 |
+
"eval_samples_per_second": 34.5,
|
| 237 |
+
"eval_steps_per_second": 4.317,
|
| 238 |
+
"eval_wer": 0.3286703632975174,
|
| 239 |
+
"step": 2300
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 4.95,
|
| 243 |
+
"eval_loss": 0.43576258420944214,
|
| 244 |
+
"eval_runtime": 111.7707,
|
| 245 |
+
"eval_samples_per_second": 34.678,
|
| 246 |
+
"eval_steps_per_second": 4.339,
|
| 247 |
+
"eval_wer": 0.3264155718657249,
|
| 248 |
+
"step": 2400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.99,
|
| 252 |
+
"step": 2420,
|
| 253 |
+
"total_flos": 8.891283011181065e+18,
|
| 254 |
+
"train_loss": 1.0930316641311015,
|
| 255 |
+
"train_runtime": 6038.6358,
|
| 256 |
+
"train_samples_per_second": 12.837,
|
| 257 |
+
"train_steps_per_second": 0.401
|
| 258 |
+
}
|
| 259 |
+
],
|
| 260 |
+
"max_steps": 2420,
|
| 261 |
+
"num_train_epochs": 5,
|
| 262 |
+
"total_flos": 8.891283011181065e+18,
|
| 263 |
+
"trial_name": null,
|
| 264 |
+
"trial_params": null
|
| 265 |
+
}
|