End of training
Browse files- all_results.json +12 -0
- eval_results.json +8 -0
- runs/Aug18_10-10-50_gpu01/events.out.tfevents.1692356548.gpu01.18551.1 +3 -0
- train_results.json +7 -0
- trainer_state.json +232 -0
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.36,
|
| 3 |
+
"eval_loss": 0.642578125,
|
| 4 |
+
"eval_runtime": 165.5744,
|
| 5 |
+
"eval_samples_per_second": 53.323,
|
| 6 |
+
"eval_steps_per_second": 3.334,
|
| 7 |
+
"eval_wer": 53.0400387724153,
|
| 8 |
+
"train_loss": 0.8101146443684896,
|
| 9 |
+
"train_runtime": 1696.2635,
|
| 10 |
+
"train_samples_per_second": 45.276,
|
| 11 |
+
"train_steps_per_second": 0.177
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.36,
|
| 3 |
+
"eval_loss": 0.642578125,
|
| 4 |
+
"eval_runtime": 165.5744,
|
| 5 |
+
"eval_samples_per_second": 53.323,
|
| 6 |
+
"eval_steps_per_second": 3.334,
|
| 7 |
+
"eval_wer": 53.0400387724153
|
| 8 |
+
}
|
runs/Aug18_10-10-50_gpu01/events.out.tfevents.1692356548.gpu01.18551.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68e7bf8ad7e1761c738066685e218b1d02fb6c93f1c0089401bf3b026027afa2
|
| 3 |
+
size 406
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.36,
|
| 3 |
+
"train_loss": 0.8101146443684896,
|
| 4 |
+
"train_runtime": 1696.2635,
|
| 5 |
+
"train_samples_per_second": 45.276,
|
| 6 |
+
"train_steps_per_second": 0.177
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,232 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 53.0400387724153,
|
| 3 |
+
"best_model_checkpoint": "./whisper_tiny_cs/checkpoint-300",
|
| 4 |
+
"epoch": 4.363636363636363,
|
| 5 |
+
"global_step": 300,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.15,
|
| 12 |
+
"learning_rate": 2.5938375012788124e-06,
|
| 13 |
+
"loss": 1.9204,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.29,
|
| 18 |
+
"learning_rate": 4.0137332755197495e-06,
|
| 19 |
+
"loss": 1.6483,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.44,
|
| 24 |
+
"learning_rate": 4.716577826569394e-06,
|
| 25 |
+
"loss": 1.3926,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.58,
|
| 30 |
+
"learning_rate": 5.187675002557625e-06,
|
| 31 |
+
"loss": 1.215,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.73,
|
| 36 |
+
"learning_rate": 5.542526105605248e-06,
|
| 37 |
+
"loss": 1.0921,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.87,
|
| 42 |
+
"learning_rate": 5.827293423354003e-06,
|
| 43 |
+
"loss": 1.0174,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.02,
|
| 48 |
+
"learning_rate": 6.065146451806229e-06,
|
| 49 |
+
"loss": 0.9591,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.16,
|
| 54 |
+
"learning_rate": 6.2693786409359715e-06,
|
| 55 |
+
"loss": 0.8963,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.31,
|
| 60 |
+
"learning_rate": 6.448328170811893e-06,
|
| 61 |
+
"loss": 0.8609,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.45,
|
| 66 |
+
"learning_rate": 6.607570776798562e-06,
|
| 67 |
+
"loss": 0.8297,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.45,
|
| 72 |
+
"eval_loss": 0.873046875,
|
| 73 |
+
"eval_runtime": 168.9595,
|
| 74 |
+
"eval_samples_per_second": 52.255,
|
| 75 |
+
"eval_steps_per_second": 3.267,
|
| 76 |
+
"eval_wer": 66.3524224942545,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.6,
|
| 81 |
+
"learning_rate": 6.751019550882568e-06,
|
| 82 |
+
"loss": 0.7988,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.75,
|
| 87 |
+
"learning_rate": 6.881526630756396e-06,
|
| 88 |
+
"loss": 0.7724,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.89,
|
| 93 |
+
"learning_rate": 7.001235150391877e-06,
|
| 94 |
+
"loss": 0.7666,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.04,
|
| 99 |
+
"learning_rate": 7.11179636123406e-06,
|
| 100 |
+
"loss": 0.7285,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.18,
|
| 105 |
+
"learning_rate": 7.214509519281458e-06,
|
| 106 |
+
"loss": 0.6919,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 2.33,
|
| 111 |
+
"learning_rate": 7.310415327848206e-06,
|
| 112 |
+
"loss": 0.6813,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.47,
|
| 117 |
+
"learning_rate": 7.400360293466852e-06,
|
| 118 |
+
"loss": 0.6646,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 2.62,
|
| 123 |
+
"learning_rate": 7.485042226047167e-06,
|
| 124 |
+
"loss": 0.6515,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.76,
|
| 129 |
+
"learning_rate": 7.565043147393057e-06,
|
| 130 |
+
"loss": 0.6322,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.91,
|
| 135 |
+
"learning_rate": 7.640853571188253e-06,
|
| 136 |
+
"loss": 0.62,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.91,
|
| 141 |
+
"eval_loss": 0.71875,
|
| 142 |
+
"eval_runtime": 169.5645,
|
| 143 |
+
"eval_samples_per_second": 52.069,
|
| 144 |
+
"eval_steps_per_second": 3.255,
|
| 145 |
+
"eval_wer": 57.866266435282895,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.05,
|
| 150 |
+
"learning_rate": 7.712890734563846e-06,
|
| 151 |
+
"loss": 0.6029,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 3.2,
|
| 156 |
+
"learning_rate": 7.781512503836438e-06,
|
| 157 |
+
"loss": 0.5771,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 3.35,
|
| 162 |
+
"learning_rate": 7.847028130491337e-06,
|
| 163 |
+
"loss": 0.5742,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.49,
|
| 168 |
+
"learning_rate": 7.909706676567024e-06,
|
| 169 |
+
"loss": 0.5557,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.64,
|
| 174 |
+
"learning_rate": 7.969783690344598e-06,
|
| 175 |
+
"loss": 0.5569,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.78,
|
| 180 |
+
"learning_rate": 8.027466551039499e-06,
|
| 181 |
+
"loss": 0.5467,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 3.93,
|
| 186 |
+
"learning_rate": 8.082938788770226e-06,
|
| 187 |
+
"loss": 0.5358,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.07,
|
| 192 |
+
"learning_rate": 8.13636360688406e-06,
|
| 193 |
+
"loss": 0.5169,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 4.22,
|
| 198 |
+
"learning_rate": 8.187886777096811e-06,
|
| 199 |
+
"loss": 0.499,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.36,
|
| 204 |
+
"learning_rate": 8.237639036863129e-06,
|
| 205 |
+
"loss": 0.4986,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 4.36,
|
| 210 |
+
"eval_loss": 0.642578125,
|
| 211 |
+
"eval_runtime": 166.9295,
|
| 212 |
+
"eval_samples_per_second": 52.891,
|
| 213 |
+
"eval_steps_per_second": 3.307,
|
| 214 |
+
"eval_wer": 53.0400387724153,
|
| 215 |
+
"step": 300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 4.36,
|
| 219 |
+
"step": 300,
|
| 220 |
+
"total_flos": 1.890730374196101e+18,
|
| 221 |
+
"train_loss": 0.8101146443684896,
|
| 222 |
+
"train_runtime": 1696.2635,
|
| 223 |
+
"train_samples_per_second": 45.276,
|
| 224 |
+
"train_steps_per_second": 0.177
|
| 225 |
+
}
|
| 226 |
+
],
|
| 227 |
+
"max_steps": 300,
|
| 228 |
+
"num_train_epochs": 5,
|
| 229 |
+
"total_flos": 1.890730374196101e+18,
|
| 230 |
+
"trial_name": null,
|
| 231 |
+
"trial_params": null
|
| 232 |
+
}
|