Commit
•
2dc0b75
1
Parent(s):
3c4fd4a
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.9,
|
| 3 |
+
"eval_accuracy": 0.9898089171974522,
|
| 4 |
+
"eval_loss": 0.05419470742344856,
|
| 5 |
+
"eval_runtime": 11.1912,
|
| 6 |
+
"eval_samples_per_second": 70.144,
|
| 7 |
+
"eval_steps_per_second": 2.949,
|
| 8 |
+
"total_flos": 5.416738622139802e+18,
|
| 9 |
+
"train_loss": 0.011492813212720498,
|
| 10 |
+
"train_runtime": 2907.1139,
|
| 11 |
+
"train_samples_per_second": 24.285,
|
| 12 |
+
"train_steps_per_second": 0.251
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.9,
|
| 3 |
+
"eval_accuracy": 0.9898089171974522,
|
| 4 |
+
"eval_loss": 0.05419470742344856,
|
| 5 |
+
"eval_runtime": 11.1912,
|
| 6 |
+
"eval_samples_per_second": 70.144,
|
| 7 |
+
"eval_steps_per_second": 2.949
|
| 8 |
+
}
|
runs/Aug11_14-34-37_fcf4b37c5947/events.out.tfevents.1691767570.fcf4b37c5947.345.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ce0480da7ea75aefb271b18b7f8d1afc3d033447686a1140fb42b1ae2bcdf396
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.9,
|
| 3 |
+
"total_flos": 5.416738622139802e+18,
|
| 4 |
+
"train_loss": 0.011492813212720498,
|
| 5 |
+
"train_runtime": 2907.1139,
|
| 6 |
+
"train_samples_per_second": 24.285,
|
| 7 |
+
"train_steps_per_second": 0.251
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,553 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9898089171974522,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-in21k-mobile-eye-tracking-dataset-v2/checkpoint-221",
|
| 4 |
+
"epoch": 9.898305084745763,
|
| 5 |
+
"global_step": 730,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.14,
|
| 12 |
+
"learning_rate": 6.849315068493151e-06,
|
| 13 |
+
"loss": 0.0366,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.27,
|
| 18 |
+
"learning_rate": 1.3698630136986302e-05,
|
| 19 |
+
"loss": 0.0367,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.41,
|
| 24 |
+
"learning_rate": 2.0547945205479453e-05,
|
| 25 |
+
"loss": 0.0341,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.54,
|
| 30 |
+
"learning_rate": 2.7397260273972603e-05,
|
| 31 |
+
"loss": 0.0396,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.68,
|
| 36 |
+
"learning_rate": 3.424657534246575e-05,
|
| 37 |
+
"loss": 0.0338,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.81,
|
| 42 |
+
"learning_rate": 4.1095890410958905e-05,
|
| 43 |
+
"loss": 0.0298,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.95,
|
| 48 |
+
"learning_rate": 4.794520547945205e-05,
|
| 49 |
+
"loss": 0.024,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.99,
|
| 54 |
+
"eval_accuracy": 0.9808917197452229,
|
| 55 |
+
"eval_loss": 0.0768975093960762,
|
| 56 |
+
"eval_runtime": 10.9295,
|
| 57 |
+
"eval_samples_per_second": 71.824,
|
| 58 |
+
"eval_steps_per_second": 3.019,
|
| 59 |
+
"step": 73
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.08,
|
| 63 |
+
"learning_rate": 4.9467275494672755e-05,
|
| 64 |
+
"loss": 0.0206,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.22,
|
| 69 |
+
"learning_rate": 4.8706240487062404e-05,
|
| 70 |
+
"loss": 0.0219,
|
| 71 |
+
"step": 90
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 1.36,
|
| 75 |
+
"learning_rate": 4.794520547945205e-05,
|
| 76 |
+
"loss": 0.0167,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.49,
|
| 81 |
+
"learning_rate": 4.71841704718417e-05,
|
| 82 |
+
"loss": 0.0314,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.63,
|
| 87 |
+
"learning_rate": 4.642313546423136e-05,
|
| 88 |
+
"loss": 0.0304,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.76,
|
| 93 |
+
"learning_rate": 4.5662100456621006e-05,
|
| 94 |
+
"loss": 0.017,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.9,
|
| 99 |
+
"learning_rate": 4.4901065449010655e-05,
|
| 100 |
+
"loss": 0.0236,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.99,
|
| 105 |
+
"eval_accuracy": 0.9745222929936306,
|
| 106 |
+
"eval_loss": 0.11105378717184067,
|
| 107 |
+
"eval_runtime": 10.6001,
|
| 108 |
+
"eval_samples_per_second": 74.056,
|
| 109 |
+
"eval_steps_per_second": 3.113,
|
| 110 |
+
"step": 147
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.03,
|
| 114 |
+
"learning_rate": 4.41400304414003e-05,
|
| 115 |
+
"loss": 0.022,
|
| 116 |
+
"step": 150
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.17,
|
| 120 |
+
"learning_rate": 4.337899543378995e-05,
|
| 121 |
+
"loss": 0.0184,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.31,
|
| 126 |
+
"learning_rate": 4.26179604261796e-05,
|
| 127 |
+
"loss": 0.0321,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.44,
|
| 132 |
+
"learning_rate": 4.1856925418569256e-05,
|
| 133 |
+
"loss": 0.015,
|
| 134 |
+
"step": 180
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.58,
|
| 138 |
+
"learning_rate": 4.1095890410958905e-05,
|
| 139 |
+
"loss": 0.0114,
|
| 140 |
+
"step": 190
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.71,
|
| 144 |
+
"learning_rate": 4.0334855403348554e-05,
|
| 145 |
+
"loss": 0.0155,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.85,
|
| 150 |
+
"learning_rate": 3.95738203957382e-05,
|
| 151 |
+
"loss": 0.0165,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.98,
|
| 156 |
+
"learning_rate": 3.881278538812785e-05,
|
| 157 |
+
"loss": 0.0172,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 3.0,
|
| 162 |
+
"eval_accuracy": 0.9898089171974522,
|
| 163 |
+
"eval_loss": 0.05419470742344856,
|
| 164 |
+
"eval_runtime": 11.1217,
|
| 165 |
+
"eval_samples_per_second": 70.583,
|
| 166 |
+
"eval_steps_per_second": 2.967,
|
| 167 |
+
"step": 221
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 3.12,
|
| 171 |
+
"learning_rate": 3.80517503805175e-05,
|
| 172 |
+
"loss": 0.0086,
|
| 173 |
+
"step": 230
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 3.25,
|
| 177 |
+
"learning_rate": 3.7290715372907156e-05,
|
| 178 |
+
"loss": 0.0145,
|
| 179 |
+
"step": 240
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 3.39,
|
| 183 |
+
"learning_rate": 3.6529680365296805e-05,
|
| 184 |
+
"loss": 0.0082,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 3.53,
|
| 189 |
+
"learning_rate": 3.5768645357686453e-05,
|
| 190 |
+
"loss": 0.0087,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 3.66,
|
| 195 |
+
"learning_rate": 3.50076103500761e-05,
|
| 196 |
+
"loss": 0.0072,
|
| 197 |
+
"step": 270
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.8,
|
| 201 |
+
"learning_rate": 3.424657534246575e-05,
|
| 202 |
+
"loss": 0.0171,
|
| 203 |
+
"step": 280
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 3.93,
|
| 207 |
+
"learning_rate": 3.34855403348554e-05,
|
| 208 |
+
"loss": 0.0114,
|
| 209 |
+
"step": 290
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 4.0,
|
| 213 |
+
"eval_accuracy": 0.9885350318471338,
|
| 214 |
+
"eval_loss": 0.06302983313798904,
|
| 215 |
+
"eval_runtime": 10.6941,
|
| 216 |
+
"eval_samples_per_second": 73.405,
|
| 217 |
+
"eval_steps_per_second": 3.086,
|
| 218 |
+
"step": 295
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.07,
|
| 222 |
+
"learning_rate": 3.2724505327245055e-05,
|
| 223 |
+
"loss": 0.0063,
|
| 224 |
+
"step": 300
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 4.2,
|
| 228 |
+
"learning_rate": 3.1963470319634704e-05,
|
| 229 |
+
"loss": 0.006,
|
| 230 |
+
"step": 310
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 4.34,
|
| 234 |
+
"learning_rate": 3.120243531202435e-05,
|
| 235 |
+
"loss": 0.0102,
|
| 236 |
+
"step": 320
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 4.47,
|
| 240 |
+
"learning_rate": 3.0441400304414e-05,
|
| 241 |
+
"loss": 0.0054,
|
| 242 |
+
"step": 330
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.61,
|
| 246 |
+
"learning_rate": 2.9680365296803654e-05,
|
| 247 |
+
"loss": 0.0074,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.75,
|
| 252 |
+
"learning_rate": 2.8919330289193303e-05,
|
| 253 |
+
"loss": 0.0108,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 4.88,
|
| 258 |
+
"learning_rate": 2.815829528158295e-05,
|
| 259 |
+
"loss": 0.0051,
|
| 260 |
+
"step": 360
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 4.99,
|
| 264 |
+
"eval_accuracy": 0.9859872611464968,
|
| 265 |
+
"eval_loss": 0.0673714205622673,
|
| 266 |
+
"eval_runtime": 10.7583,
|
| 267 |
+
"eval_samples_per_second": 72.967,
|
| 268 |
+
"eval_steps_per_second": 3.067,
|
| 269 |
+
"step": 368
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 5.02,
|
| 273 |
+
"learning_rate": 2.7397260273972603e-05,
|
| 274 |
+
"loss": 0.005,
|
| 275 |
+
"step": 370
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 5.15,
|
| 279 |
+
"learning_rate": 2.6636225266362252e-05,
|
| 280 |
+
"loss": 0.0048,
|
| 281 |
+
"step": 380
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 5.29,
|
| 285 |
+
"learning_rate": 2.58751902587519e-05,
|
| 286 |
+
"loss": 0.0084,
|
| 287 |
+
"step": 390
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 5.42,
|
| 291 |
+
"learning_rate": 2.5114155251141553e-05,
|
| 292 |
+
"loss": 0.0049,
|
| 293 |
+
"step": 400
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 5.56,
|
| 297 |
+
"learning_rate": 2.4353120243531202e-05,
|
| 298 |
+
"loss": 0.0045,
|
| 299 |
+
"step": 410
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 5.69,
|
| 303 |
+
"learning_rate": 2.359208523592085e-05,
|
| 304 |
+
"loss": 0.0043,
|
| 305 |
+
"step": 420
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 5.83,
|
| 309 |
+
"learning_rate": 2.2831050228310503e-05,
|
| 310 |
+
"loss": 0.0097,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 5.97,
|
| 315 |
+
"learning_rate": 2.207001522070015e-05,
|
| 316 |
+
"loss": 0.0044,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 5.99,
|
| 321 |
+
"eval_accuracy": 0.9885350318471338,
|
| 322 |
+
"eval_loss": 0.06396039575338364,
|
| 323 |
+
"eval_runtime": 10.7469,
|
| 324 |
+
"eval_samples_per_second": 73.044,
|
| 325 |
+
"eval_steps_per_second": 3.071,
|
| 326 |
+
"step": 442
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 6.1,
|
| 330 |
+
"learning_rate": 2.13089802130898e-05,
|
| 331 |
+
"loss": 0.0053,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 6.24,
|
| 336 |
+
"learning_rate": 2.0547945205479453e-05,
|
| 337 |
+
"loss": 0.0041,
|
| 338 |
+
"step": 460
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 6.37,
|
| 342 |
+
"learning_rate": 1.97869101978691e-05,
|
| 343 |
+
"loss": 0.004,
|
| 344 |
+
"step": 470
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 6.51,
|
| 348 |
+
"learning_rate": 1.902587519025875e-05,
|
| 349 |
+
"loss": 0.004,
|
| 350 |
+
"step": 480
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 6.64,
|
| 354 |
+
"learning_rate": 1.8264840182648402e-05,
|
| 355 |
+
"loss": 0.0039,
|
| 356 |
+
"step": 490
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 6.78,
|
| 360 |
+
"learning_rate": 1.750380517503805e-05,
|
| 361 |
+
"loss": 0.0091,
|
| 362 |
+
"step": 500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 6.92,
|
| 366 |
+
"learning_rate": 1.67427701674277e-05,
|
| 367 |
+
"loss": 0.0037,
|
| 368 |
+
"step": 510
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 7.0,
|
| 372 |
+
"eval_accuracy": 0.9885350318471338,
|
| 373 |
+
"eval_loss": 0.06461471319198608,
|
| 374 |
+
"eval_runtime": 10.6595,
|
| 375 |
+
"eval_samples_per_second": 73.643,
|
| 376 |
+
"eval_steps_per_second": 3.096,
|
| 377 |
+
"step": 516
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 7.05,
|
| 381 |
+
"learning_rate": 1.5981735159817352e-05,
|
| 382 |
+
"loss": 0.0085,
|
| 383 |
+
"step": 520
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 7.19,
|
| 387 |
+
"learning_rate": 1.5220700152207e-05,
|
| 388 |
+
"loss": 0.0037,
|
| 389 |
+
"step": 530
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 7.32,
|
| 393 |
+
"learning_rate": 1.4459665144596651e-05,
|
| 394 |
+
"loss": 0.0036,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 7.46,
|
| 399 |
+
"learning_rate": 1.3698630136986302e-05,
|
| 400 |
+
"loss": 0.0036,
|
| 401 |
+
"step": 550
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 7.59,
|
| 405 |
+
"learning_rate": 1.293759512937595e-05,
|
| 406 |
+
"loss": 0.0035,
|
| 407 |
+
"step": 560
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 7.73,
|
| 411 |
+
"learning_rate": 1.2176560121765601e-05,
|
| 412 |
+
"loss": 0.0035,
|
| 413 |
+
"step": 570
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 7.86,
|
| 417 |
+
"learning_rate": 1.1415525114155251e-05,
|
| 418 |
+
"loss": 0.0034,
|
| 419 |
+
"step": 580
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 8.0,
|
| 423 |
+
"learning_rate": 1.06544901065449e-05,
|
| 424 |
+
"loss": 0.0034,
|
| 425 |
+
"step": 590
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 8.0,
|
| 429 |
+
"eval_accuracy": 0.9885350318471338,
|
| 430 |
+
"eval_loss": 0.06516698002815247,
|
| 431 |
+
"eval_runtime": 10.6581,
|
| 432 |
+
"eval_samples_per_second": 73.653,
|
| 433 |
+
"eval_steps_per_second": 3.096,
|
| 434 |
+
"step": 590
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 8.14,
|
| 438 |
+
"learning_rate": 9.89345509893455e-06,
|
| 439 |
+
"loss": 0.0034,
|
| 440 |
+
"step": 600
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 8.27,
|
| 444 |
+
"learning_rate": 9.132420091324201e-06,
|
| 445 |
+
"loss": 0.0033,
|
| 446 |
+
"step": 610
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 8.41,
|
| 450 |
+
"learning_rate": 8.37138508371385e-06,
|
| 451 |
+
"loss": 0.0033,
|
| 452 |
+
"step": 620
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 8.54,
|
| 456 |
+
"learning_rate": 7.6103500761035e-06,
|
| 457 |
+
"loss": 0.0033,
|
| 458 |
+
"step": 630
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 8.68,
|
| 462 |
+
"learning_rate": 6.849315068493151e-06,
|
| 463 |
+
"loss": 0.008,
|
| 464 |
+
"step": 640
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 8.81,
|
| 468 |
+
"learning_rate": 6.0882800608828005e-06,
|
| 469 |
+
"loss": 0.0033,
|
| 470 |
+
"step": 650
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 8.95,
|
| 474 |
+
"learning_rate": 5.32724505327245e-06,
|
| 475 |
+
"loss": 0.0032,
|
| 476 |
+
"step": 660
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 8.99,
|
| 480 |
+
"eval_accuracy": 0.9885350318471338,
|
| 481 |
+
"eval_loss": 0.06558429449796677,
|
| 482 |
+
"eval_runtime": 10.7487,
|
| 483 |
+
"eval_samples_per_second": 73.032,
|
| 484 |
+
"eval_steps_per_second": 3.07,
|
| 485 |
+
"step": 663
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 9.08,
|
| 489 |
+
"learning_rate": 4.566210045662101e-06,
|
| 490 |
+
"loss": 0.0031,
|
| 491 |
+
"step": 670
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 9.22,
|
| 495 |
+
"learning_rate": 3.80517503805175e-06,
|
| 496 |
+
"loss": 0.0033,
|
| 497 |
+
"step": 680
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 9.36,
|
| 501 |
+
"learning_rate": 3.0441400304414002e-06,
|
| 502 |
+
"loss": 0.0032,
|
| 503 |
+
"step": 690
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 9.49,
|
| 507 |
+
"learning_rate": 2.2831050228310503e-06,
|
| 508 |
+
"loss": 0.0031,
|
| 509 |
+
"step": 700
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 9.63,
|
| 513 |
+
"learning_rate": 1.5220700152207001e-06,
|
| 514 |
+
"loss": 0.0032,
|
| 515 |
+
"step": 710
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 9.76,
|
| 519 |
+
"learning_rate": 7.610350076103501e-07,
|
| 520 |
+
"loss": 0.0076,
|
| 521 |
+
"step": 720
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 9.9,
|
| 525 |
+
"learning_rate": 0.0,
|
| 526 |
+
"loss": 0.0032,
|
| 527 |
+
"step": 730
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 9.9,
|
| 531 |
+
"eval_accuracy": 0.9885350318471338,
|
| 532 |
+
"eval_loss": 0.06568808108568192,
|
| 533 |
+
"eval_runtime": 10.6014,
|
| 534 |
+
"eval_samples_per_second": 74.047,
|
| 535 |
+
"eval_steps_per_second": 3.113,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 9.9,
|
| 540 |
+
"step": 730,
|
| 541 |
+
"total_flos": 5.416738622139802e+18,
|
| 542 |
+
"train_loss": 0.011492813212720498,
|
| 543 |
+
"train_runtime": 2907.1139,
|
| 544 |
+
"train_samples_per_second": 24.285,
|
| 545 |
+
"train_steps_per_second": 0.251
|
| 546 |
+
}
|
| 547 |
+
],
|
| 548 |
+
"max_steps": 730,
|
| 549 |
+
"num_train_epochs": 10,
|
| 550 |
+
"total_flos": 5.416738622139802e+18,
|
| 551 |
+
"trial_name": null,
|
| 552 |
+
"trial_params": null
|
| 553 |
+
}
|