Commit
•
578781f
1
Parent(s):
46a4856
End of training
Browse files- all_results.json +14 -0
- eval_results.json +9 -0
- train_results.json +8 -0
- trainer_state.json +244 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.9693832397460938,
|
| 4 |
+
"eval_runtime": 412.1508,
|
| 5 |
+
"eval_samples": 10388,
|
| 6 |
+
"eval_samples_per_second": 25.204,
|
| 7 |
+
"eval_steps_per_second": 3.152,
|
| 8 |
+
"eval_wer": 0.7823963593984128,
|
| 9 |
+
"train_loss": 2.3014913469352978,
|
| 10 |
+
"train_runtime": 21335.6515,
|
| 11 |
+
"train_samples": 38209,
|
| 12 |
+
"train_samples_per_second": 17.909,
|
| 13 |
+
"train_steps_per_second": 0.14
|
| 14 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 0.9693832397460938,
|
| 4 |
+
"eval_runtime": 412.1508,
|
| 5 |
+
"eval_samples": 10388,
|
| 6 |
+
"eval_samples_per_second": 25.204,
|
| 7 |
+
"eval_steps_per_second": 3.152,
|
| 8 |
+
"eval_wer": 0.7823963593984128
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 2.3014913469352978,
|
| 4 |
+
"train_runtime": 21335.6515,
|
| 5 |
+
"train_samples": 38209,
|
| 6 |
+
"train_samples_per_second": 17.909,
|
| 7 |
+
"train_steps_per_second": 0.14
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,244 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.997489539748955,
|
| 5 |
+
"global_step": 2980,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.33,
|
| 12 |
+
"learning_rate": 4.9000000000000005e-05,
|
| 13 |
+
"loss": 5.3579,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.67,
|
| 18 |
+
"learning_rate": 9.900000000000001e-05,
|
| 19 |
+
"loss": 3.2697,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.01,
|
| 24 |
+
"learning_rate": 0.000149,
|
| 25 |
+
"loss": 2.4711,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 1.34,
|
| 30 |
+
"learning_rate": 0.000199,
|
| 31 |
+
"loss": 2.1202,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.68,
|
| 36 |
+
"learning_rate": 0.000249,
|
| 37 |
+
"loss": 2.0846,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.68,
|
| 42 |
+
"eval_loss": 1.1640619039535522,
|
| 43 |
+
"eval_runtime": 414.2546,
|
| 44 |
+
"eval_samples_per_second": 25.076,
|
| 45 |
+
"eval_steps_per_second": 3.136,
|
| 46 |
+
"eval_wer": 0.8072035074088462,
|
| 47 |
+
"step": 500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 2.01,
|
| 51 |
+
"learning_rate": 0.000299,
|
| 52 |
+
"loss": 2.1202,
|
| 53 |
+
"step": 600
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.35,
|
| 57 |
+
"learning_rate": 0.00034899999999999997,
|
| 58 |
+
"loss": 2.1037,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.68,
|
| 63 |
+
"learning_rate": 0.00039900000000000005,
|
| 64 |
+
"loss": 2.0882,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 3.02,
|
| 69 |
+
"learning_rate": 0.000449,
|
| 70 |
+
"loss": 2.126,
|
| 71 |
+
"step": 900
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 3.35,
|
| 75 |
+
"learning_rate": 0.000499,
|
| 76 |
+
"loss": 2.1201,
|
| 77 |
+
"step": 1000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 3.35,
|
| 81 |
+
"eval_loss": 1.1776171922683716,
|
| 82 |
+
"eval_runtime": 424.2371,
|
| 83 |
+
"eval_samples_per_second": 24.486,
|
| 84 |
+
"eval_steps_per_second": 3.062,
|
| 85 |
+
"eval_wer": 0.8329171060177221,
|
| 86 |
+
"step": 1000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 3.69,
|
| 90 |
+
"learning_rate": 0.000549,
|
| 91 |
+
"loss": 2.1435,
|
| 92 |
+
"step": 1100
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 4.03,
|
| 96 |
+
"learning_rate": 0.000599,
|
| 97 |
+
"loss": 2.152,
|
| 98 |
+
"step": 1200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 4.36,
|
| 102 |
+
"learning_rate": 0.0006490000000000001,
|
| 103 |
+
"loss": 2.11,
|
| 104 |
+
"step": 1300
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 4.7,
|
| 108 |
+
"learning_rate": 0.000699,
|
| 109 |
+
"loss": 2.1503,
|
| 110 |
+
"step": 1400
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 5.03,
|
| 114 |
+
"learning_rate": 0.000749,
|
| 115 |
+
"loss": 2.1972,
|
| 116 |
+
"step": 1500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 5.03,
|
| 120 |
+
"eval_loss": 1.2631869316101074,
|
| 121 |
+
"eval_runtime": 410.677,
|
| 122 |
+
"eval_samples_per_second": 25.295,
|
| 123 |
+
"eval_steps_per_second": 3.163,
|
| 124 |
+
"eval_wer": 0.8723754555376732,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.37,
|
| 129 |
+
"learning_rate": 0.000799,
|
| 130 |
+
"loss": 2.1788,
|
| 131 |
+
"step": 1600
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 5.7,
|
| 135 |
+
"learning_rate": 0.000849,
|
| 136 |
+
"loss": 2.1774,
|
| 137 |
+
"step": 1700
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 6.04,
|
| 141 |
+
"learning_rate": 0.0008990000000000001,
|
| 142 |
+
"loss": 2.2205,
|
| 143 |
+
"step": 1800
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.37,
|
| 147 |
+
"learning_rate": 0.000949,
|
| 148 |
+
"loss": 2.2466,
|
| 149 |
+
"step": 1900
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 6.71,
|
| 153 |
+
"learning_rate": 0.000999,
|
| 154 |
+
"loss": 2.2643,
|
| 155 |
+
"step": 2000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 6.71,
|
| 159 |
+
"eval_loss": 1.372324824333191,
|
| 160 |
+
"eval_runtime": 411.6031,
|
| 161 |
+
"eval_samples_per_second": 25.238,
|
| 162 |
+
"eval_steps_per_second": 3.156,
|
| 163 |
+
"eval_wer": 0.8982740440645984,
|
| 164 |
+
"step": 2000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 7.05,
|
| 168 |
+
"learning_rate": 0.0009000000000000001,
|
| 169 |
+
"loss": 2.2996,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 7.38,
|
| 174 |
+
"learning_rate": 0.0007979591836734694,
|
| 175 |
+
"loss": 2.2775,
|
| 176 |
+
"step": 2200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 7.72,
|
| 180 |
+
"learning_rate": 0.0006959183673469388,
|
| 181 |
+
"loss": 2.2478,
|
| 182 |
+
"step": 2300
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 8.05,
|
| 186 |
+
"learning_rate": 0.0005938775510204082,
|
| 187 |
+
"loss": 2.2288,
|
| 188 |
+
"step": 2400
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 8.39,
|
| 192 |
+
"learning_rate": 0.0004918367346938776,
|
| 193 |
+
"loss": 2.1649,
|
| 194 |
+
"step": 2500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 8.39,
|
| 198 |
+
"eval_loss": 1.25503671169281,
|
| 199 |
+
"eval_runtime": 416.24,
|
| 200 |
+
"eval_samples_per_second": 24.957,
|
| 201 |
+
"eval_steps_per_second": 3.121,
|
| 202 |
+
"eval_wer": 0.884196311301034,
|
| 203 |
+
"step": 2500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 8.72,
|
| 207 |
+
"learning_rate": 0.000389795918367347,
|
| 208 |
+
"loss": 2.1054,
|
| 209 |
+
"step": 2600
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.06,
|
| 213 |
+
"learning_rate": 0.00028775510204081633,
|
| 214 |
+
"loss": 2.0872,
|
| 215 |
+
"step": 2700
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 9.39,
|
| 219 |
+
"learning_rate": 0.00018571428571428572,
|
| 220 |
+
"loss": 1.9953,
|
| 221 |
+
"step": 2800
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 9.73,
|
| 225 |
+
"learning_rate": 8.36734693877551e-05,
|
| 226 |
+
"loss": 1.9474,
|
| 227 |
+
"step": 2900
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 10.0,
|
| 231 |
+
"step": 2980,
|
| 232 |
+
"total_flos": 1.006981652948512e+20,
|
| 233 |
+
"train_loss": 2.3014913469352978,
|
| 234 |
+
"train_runtime": 21335.6515,
|
| 235 |
+
"train_samples_per_second": 17.909,
|
| 236 |
+
"train_steps_per_second": 0.14
|
| 237 |
+
}
|
| 238 |
+
],
|
| 239 |
+
"max_steps": 2980,
|
| 240 |
+
"num_train_epochs": 10,
|
| 241 |
+
"total_flos": 1.006981652948512e+20,
|
| 242 |
+
"trial_name": null,
|
| 243 |
+
"trial_params": null
|
| 244 |
+
}
|