Training in progress, epoch 1
Browse files- all_results.json +10 -10
- config.json +1 -3
- eval_results.json +6 -6
- pytorch_model.bin +2 -2
- train_results.json +5 -5
- trainer_state.json +1357 -379
- training_args.bin +1 -1
all_results.json
CHANGED
|
@@ -1,12 +1,12 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
-
"train_loss": 0.
|
| 9 |
-
"train_runtime":
|
| 10 |
-
"train_samples_per_second":
|
| 11 |
-
"train_steps_per_second": 0.
|
| 12 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.8832997987927566,
|
| 4 |
+
"eval_loss": 0.2902894914150238,
|
| 5 |
+
"eval_runtime": 9.9967,
|
| 6 |
+
"eval_samples_per_second": 49.716,
|
| 7 |
+
"eval_steps_per_second": 1.601,
|
| 8 |
+
"train_loss": 0.044299079362735416,
|
| 9 |
+
"train_runtime": 9406.6232,
|
| 10 |
+
"train_samples_per_second": 23.755,
|
| 11 |
+
"train_steps_per_second": 0.186
|
| 12 |
}
|
config.json
CHANGED
|
@@ -19,14 +19,12 @@
|
|
| 19 |
"id2label": {
|
| 20 |
"0": "castravete",
|
| 21 |
"1": "rosie",
|
| 22 |
-
"2": "salata"
|
| 23 |
-
"3": "ridiche"
|
| 24 |
},
|
| 25 |
"image_size": 224,
|
| 26 |
"initializer_range": 0.02,
|
| 27 |
"label2id": {
|
| 28 |
"castravete": 0,
|
| 29 |
-
"ridiche": 3,
|
| 30 |
"rosie": 1,
|
| 31 |
"salata": 2
|
| 32 |
},
|
|
|
|
| 19 |
"id2label": {
|
| 20 |
"0": "castravete",
|
| 21 |
"1": "rosie",
|
| 22 |
+
"2": "salata"
|
|
|
|
| 23 |
},
|
| 24 |
"image_size": 224,
|
| 25 |
"initializer_range": 0.02,
|
| 26 |
"label2id": {
|
| 27 |
"castravete": 0,
|
|
|
|
| 28 |
"rosie": 1,
|
| 29 |
"salata": 2
|
| 30 |
},
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"eval_accuracy": 0.8832997987927566,
|
| 4 |
+
"eval_loss": 0.2902894914150238,
|
| 5 |
+
"eval_runtime": 9.9967,
|
| 6 |
+
"eval_samples_per_second": 49.716,
|
| 7 |
+
"eval_steps_per_second": 1.601
|
| 8 |
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d58f9acf59892f6d432f6c6d3ebccd02a79cdd4a5573dec67a4c2050b04b5f3c
|
| 3 |
+
size 110395951
|
train_results.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss": 0.
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples_per_second":
|
| 6 |
-
"train_steps_per_second": 0.
|
| 7 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 50.0,
|
| 3 |
+
"train_loss": 0.044299079362735416,
|
| 4 |
+
"train_runtime": 9406.6232,
|
| 5 |
+
"train_samples_per_second": 23.755,
|
| 6 |
+
"train_steps_per_second": 0.186
|
| 7 |
}
|
trainer_state.json
CHANGED
|
@@ -1,547 +1,1525 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-agrivision/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 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.
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss": 1.
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"
|
| 19 |
-
"
|
| 20 |
-
"eval_runtime": 4.1559,
|
| 21 |
-
"eval_samples_per_second": 48.605,
|
| 22 |
-
"eval_steps_per_second": 1.684,
|
| 23 |
-
"step": 14
|
| 24 |
-
},
|
| 25 |
-
{
|
| 26 |
-
"epoch": 1.42,
|
| 27 |
-
"learning_rate": 2.380952380952381e-05,
|
| 28 |
-
"loss": 0.9466,
|
| 29 |
"step": 20
|
| 30 |
},
|
| 31 |
{
|
| 32 |
-
"epoch":
|
| 33 |
-
"
|
| 34 |
-
"
|
| 35 |
-
"eval_runtime": 4.2775,
|
| 36 |
-
"eval_samples_per_second": 47.224,
|
| 37 |
-
"eval_steps_per_second": 1.636,
|
| 38 |
-
"step": 28
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 2.14,
|
| 42 |
-
"learning_rate": 3.571428571428572e-05,
|
| 43 |
-
"loss": 0.5439,
|
| 44 |
"step": 30
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch":
|
| 48 |
-
"
|
| 49 |
-
"
|
| 50 |
-
"
|
|
|
|
|
|
|
|
|
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"epoch":
|
| 54 |
-
"
|
| 55 |
-
"
|
| 56 |
-
"
|
| 57 |
-
"eval_samples_per_second": 49.092,
|
| 58 |
-
"eval_steps_per_second": 1.701,
|
| 59 |
-
"step": 42
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 50
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch":
|
| 69 |
-
"
|
| 70 |
-
"
|
| 71 |
-
"eval_runtime": 4.1151,
|
| 72 |
-
"eval_samples_per_second": 49.088,
|
| 73 |
-
"eval_steps_per_second": 1.701,
|
| 74 |
-
"step": 56
|
| 75 |
-
},
|
| 76 |
-
{
|
| 77 |
-
"epoch": 4.28,
|
| 78 |
-
"learning_rate": 4.761904761904762e-05,
|
| 79 |
-
"loss": 0.1359,
|
| 80 |
"step": 60
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch":
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 0.
|
| 86 |
"step": 70
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"epoch":
|
| 90 |
-
"eval_accuracy": 0.
|
| 91 |
-
"eval_loss": 0.
|
| 92 |
-
"eval_runtime":
|
| 93 |
-
"eval_samples_per_second": 49.
|
| 94 |
-
"eval_steps_per_second": 1.
|
| 95 |
"step": 70
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"epoch":
|
| 99 |
-
"learning_rate":
|
| 100 |
-
"loss": 0.
|
| 101 |
"step": 80
|
| 102 |
},
|
| 103 |
{
|
| 104 |
-
"epoch":
|
| 105 |
-
"
|
| 106 |
-
"
|
| 107 |
-
"eval_runtime": 4.093,
|
| 108 |
-
"eval_samples_per_second": 49.353,
|
| 109 |
-
"eval_steps_per_second": 1.71,
|
| 110 |
-
"step": 84
|
| 111 |
-
},
|
| 112 |
-
{
|
| 113 |
-
"epoch": 6.42,
|
| 114 |
-
"learning_rate": 4.3650793650793655e-05,
|
| 115 |
-
"loss": 0.044,
|
| 116 |
"step": 90
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch":
|
| 120 |
-
"
|
| 121 |
-
"
|
| 122 |
-
"eval_runtime": 4.0736,
|
| 123 |
-
"eval_samples_per_second": 49.587,
|
| 124 |
-
"eval_steps_per_second": 1.718,
|
| 125 |
-
"step": 98
|
| 126 |
-
},
|
| 127 |
-
{
|
| 128 |
-
"epoch": 7.14,
|
| 129 |
-
"learning_rate": 4.232804232804233e-05,
|
| 130 |
-
"loss": 0.0575,
|
| 131 |
"step": 100
|
| 132 |
},
|
| 133 |
{
|
| 134 |
-
"epoch":
|
| 135 |
-
"
|
| 136 |
-
"
|
| 137 |
-
"
|
|
|
|
|
|
|
|
|
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch":
|
| 141 |
-
"
|
| 142 |
-
"
|
| 143 |
-
"
|
| 144 |
-
"eval_samples_per_second": 50.672,
|
| 145 |
-
"eval_steps_per_second": 1.756,
|
| 146 |
-
"step": 112
|
| 147 |
},
|
| 148 |
{
|
| 149 |
-
"epoch":
|
| 150 |
-
"learning_rate": 3.
|
| 151 |
-
"loss": 0.
|
| 152 |
"step": 120
|
| 153 |
},
|
| 154 |
{
|
| 155 |
-
"epoch":
|
| 156 |
-
"
|
| 157 |
-
"
|
| 158 |
-
"eval_runtime": 4.0118,
|
| 159 |
-
"eval_samples_per_second": 50.352,
|
| 160 |
-
"eval_steps_per_second": 1.745,
|
| 161 |
-
"step": 126
|
| 162 |
-
},
|
| 163 |
-
{
|
| 164 |
-
"epoch": 9.28,
|
| 165 |
-
"learning_rate": 3.835978835978836e-05,
|
| 166 |
-
"loss": 0.0119,
|
| 167 |
"step": 130
|
| 168 |
},
|
| 169 |
{
|
| 170 |
-
"epoch":
|
| 171 |
-
"learning_rate":
|
| 172 |
-
"loss": 0.
|
| 173 |
"step": 140
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch":
|
| 177 |
-
"eval_accuracy": 0.
|
| 178 |
-
"eval_loss": 0.
|
| 179 |
-
"eval_runtime":
|
| 180 |
-
"eval_samples_per_second":
|
| 181 |
-
"eval_steps_per_second": 1.
|
| 182 |
"step": 140
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 150
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch":
|
| 192 |
-
"
|
| 193 |
-
"
|
| 194 |
-
"eval_runtime": 4.0847,
|
| 195 |
-
"eval_samples_per_second": 49.452,
|
| 196 |
-
"eval_steps_per_second": 1.714,
|
| 197 |
-
"step": 154
|
| 198 |
-
},
|
| 199 |
-
{
|
| 200 |
-
"epoch": 11.42,
|
| 201 |
-
"learning_rate": 3.439153439153439e-05,
|
| 202 |
-
"loss": 0.0182,
|
| 203 |
"step": 160
|
| 204 |
},
|
| 205 |
{
|
| 206 |
-
"epoch":
|
| 207 |
-
"
|
| 208 |
-
"
|
| 209 |
-
"eval_runtime": 3.9613,
|
| 210 |
-
"eval_samples_per_second": 50.994,
|
| 211 |
-
"eval_steps_per_second": 1.767,
|
| 212 |
-
"step": 168
|
| 213 |
-
},
|
| 214 |
-
{
|
| 215 |
-
"epoch": 12.14,
|
| 216 |
-
"learning_rate": 3.306878306878307e-05,
|
| 217 |
-
"loss": 0.0086,
|
| 218 |
"step": 170
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch":
|
| 222 |
-
"
|
| 223 |
-
"
|
| 224 |
-
"
|
|
|
|
|
|
|
|
|
|
| 225 |
},
|
| 226 |
{
|
| 227 |
-
"epoch":
|
| 228 |
-
"
|
| 229 |
-
"
|
| 230 |
-
"
|
| 231 |
-
"eval_samples_per_second": 50.239,
|
| 232 |
-
"eval_steps_per_second": 1.741,
|
| 233 |
-
"step": 182
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch":
|
| 237 |
-
"learning_rate":
|
| 238 |
-
"loss": 0.
|
| 239 |
"step": 190
|
| 240 |
},
|
| 241 |
{
|
| 242 |
-
"epoch":
|
| 243 |
-
"
|
| 244 |
-
"
|
| 245 |
-
"eval_runtime": 4.056,
|
| 246 |
-
"eval_samples_per_second": 49.803,
|
| 247 |
-
"eval_steps_per_second": 1.726,
|
| 248 |
-
"step": 196
|
| 249 |
-
},
|
| 250 |
-
{
|
| 251 |
-
"epoch": 14.28,
|
| 252 |
-
"learning_rate": 2.91005291005291e-05,
|
| 253 |
-
"loss": 0.0069,
|
| 254 |
"step": 200
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch":
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 0.
|
| 260 |
"step": 210
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch":
|
| 264 |
-
"eval_accuracy": 0.
|
| 265 |
-
"eval_loss": 0.
|
| 266 |
-
"eval_runtime":
|
| 267 |
-
"eval_samples_per_second":
|
| 268 |
-
"eval_steps_per_second": 1.
|
| 269 |
"step": 210
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"epoch":
|
| 273 |
-
"learning_rate":
|
| 274 |
-
"loss": 0.
|
| 275 |
"step": 220
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch":
|
| 279 |
-
"
|
| 280 |
-
"
|
| 281 |
-
"
|
| 282 |
-
|
| 283 |
-
|
| 284 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch":
|
| 288 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 289 |
"loss": 0.0052,
|
| 290 |
-
"step":
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch":
|
| 294 |
-
"
|
| 295 |
-
"
|
| 296 |
-
"
|
| 297 |
-
|
| 298 |
-
|
| 299 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 300 |
},
|
| 301 |
{
|
| 302 |
"epoch": 17.14,
|
| 303 |
-
"learning_rate":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 304 |
"loss": 0.003,
|
| 305 |
-
"step":
|
| 306 |
},
|
| 307 |
{
|
| 308 |
-
"epoch":
|
| 309 |
-
"
|
| 310 |
-
"
|
| 311 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 312 |
},
|
| 313 |
{
|
| 314 |
-
"epoch":
|
| 315 |
-
"
|
| 316 |
-
"
|
| 317 |
-
"
|
| 318 |
-
|
| 319 |
-
|
| 320 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 321 |
},
|
| 322 |
{
|
| 323 |
-
"epoch":
|
| 324 |
-
"learning_rate":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 325 |
"loss": 0.0058,
|
| 326 |
-
"step":
|
| 327 |
},
|
| 328 |
{
|
| 329 |
-
"epoch":
|
| 330 |
-
"
|
| 331 |
-
"
|
| 332 |
-
"
|
| 333 |
-
"eval_samples_per_second": 50.22,
|
| 334 |
-
"eval_steps_per_second": 1.74,
|
| 335 |
-
"step": 266
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch":
|
| 339 |
-
"learning_rate":
|
| 340 |
-
"loss": 0.
|
| 341 |
-
"step":
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch":
|
| 345 |
-
"
|
| 346 |
-
"
|
| 347 |
-
"
|
|
|
|
|
|
|
|
|
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch":
|
| 351 |
-
"
|
| 352 |
-
"
|
| 353 |
-
"
|
| 354 |
-
"eval_samples_per_second": 50.134,
|
| 355 |
-
"eval_steps_per_second": 1.737,
|
| 356 |
-
"step": 280
|
| 357 |
},
|
| 358 |
{
|
| 359 |
-
"epoch":
|
| 360 |
-
"learning_rate":
|
| 361 |
-
"loss": 0.
|
| 362 |
-
"step":
|
| 363 |
},
|
| 364 |
{
|
| 365 |
-
"epoch":
|
| 366 |
-
"
|
| 367 |
-
"
|
| 368 |
-
"
|
| 369 |
-
"eval_samples_per_second": 49.76,
|
| 370 |
-
"eval_steps_per_second": 1.724,
|
| 371 |
-
"step": 294
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch":
|
| 375 |
-
"
|
| 376 |
-
"
|
| 377 |
-
"
|
|
|
|
|
|
|
|
|
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch":
|
| 381 |
-
"
|
| 382 |
-
"
|
| 383 |
-
"
|
| 384 |
-
"eval_samples_per_second": 51.025,
|
| 385 |
-
"eval_steps_per_second": 1.768,
|
| 386 |
-
"step": 308
|
| 387 |
},
|
| 388 |
{
|
| 389 |
-
"epoch":
|
| 390 |
-
"learning_rate":
|
| 391 |
-
"loss": 0.
|
| 392 |
-
"step":
|
| 393 |
},
|
| 394 |
{
|
| 395 |
-
"epoch":
|
| 396 |
-
"learning_rate":
|
| 397 |
-
"loss": 0.
|
| 398 |
-
"step":
|
| 399 |
},
|
| 400 |
{
|
| 401 |
-
"epoch":
|
| 402 |
-
"
|
| 403 |
-
"
|
| 404 |
-
"
|
| 405 |
-
"eval_samples_per_second": 49.943,
|
| 406 |
-
"eval_steps_per_second": 1.731,
|
| 407 |
-
"step": 322
|
| 408 |
},
|
| 409 |
{
|
| 410 |
-
"epoch":
|
| 411 |
-
"
|
| 412 |
-
"
|
| 413 |
-
"
|
|
|
|
|
|
|
|
|
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch":
|
| 417 |
-
"
|
| 418 |
-
"
|
| 419 |
-
"
|
| 420 |
-
"eval_samples_per_second": 50.137,
|
| 421 |
-
"eval_steps_per_second": 1.737,
|
| 422 |
-
"step": 336
|
| 423 |
},
|
| 424 |
{
|
| 425 |
-
"epoch":
|
| 426 |
-
"learning_rate":
|
| 427 |
-
"loss": 0.
|
| 428 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 429 |
},
|
| 430 |
{
|
| 431 |
-
"epoch":
|
| 432 |
-
"learning_rate":
|
| 433 |
"loss": 0.0008,
|
| 434 |
-
"step":
|
| 435 |
},
|
| 436 |
{
|
| 437 |
-
"epoch":
|
| 438 |
-
"
|
| 439 |
-
"
|
| 440 |
-
"
|
| 441 |
-
"eval_samples_per_second": 49.877,
|
| 442 |
-
"eval_steps_per_second": 1.728,
|
| 443 |
-
"step": 350
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"epoch":
|
| 447 |
-
"learning_rate":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 448 |
"loss": 0.0013,
|
| 449 |
-
"step":
|
| 450 |
},
|
| 451 |
{
|
| 452 |
-
"epoch":
|
| 453 |
-
"
|
| 454 |
-
"
|
| 455 |
-
"
|
| 456 |
-
"eval_samples_per_second": 49.965,
|
| 457 |
-
"eval_steps_per_second": 1.731,
|
| 458 |
-
"step": 364
|
| 459 |
},
|
| 460 |
{
|
| 461 |
-
"epoch":
|
| 462 |
-
"
|
| 463 |
-
"
|
| 464 |
-
"
|
|
|
|
|
|
|
|
|
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch":
|
| 468 |
-
"
|
| 469 |
-
"
|
| 470 |
-
"
|
| 471 |
-
"eval_samples_per_second": 50.633,
|
| 472 |
-
"eval_steps_per_second": 1.755,
|
| 473 |
-
"step": 378
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"epoch":
|
| 477 |
-
"learning_rate":
|
| 478 |
-
"loss": 0.
|
| 479 |
-
"step":
|
| 480 |
},
|
| 481 |
{
|
| 482 |
-
"epoch":
|
| 483 |
-
"learning_rate":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 484 |
"loss": 0.0019,
|
| 485 |
-
"step":
|
| 486 |
},
|
| 487 |
{
|
| 488 |
-
"epoch":
|
| 489 |
-
"eval_accuracy": 0.
|
| 490 |
-
"eval_loss": 0.
|
| 491 |
-
"eval_runtime":
|
| 492 |
-
"eval_samples_per_second":
|
| 493 |
-
"eval_steps_per_second": 1.
|
| 494 |
-
"step":
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"learning_rate":
|
| 499 |
-
"loss": 0.
|
| 500 |
-
"step":
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"epoch":
|
| 504 |
-
"
|
| 505 |
-
"
|
| 506 |
-
"
|
| 507 |
-
"eval_samples_per_second": 50.771,
|
| 508 |
-
"eval_steps_per_second": 1.759,
|
| 509 |
-
"step": 406
|
| 510 |
},
|
| 511 |
{
|
| 512 |
-
"epoch":
|
| 513 |
-
"learning_rate": 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 514 |
"loss": 0.0006,
|
| 515 |
-
"step":
|
| 516 |
},
|
| 517 |
{
|
| 518 |
-
"epoch":
|
| 519 |
-
"learning_rate":
|
| 520 |
-
"loss": 0.
|
| 521 |
-
"step":
|
| 522 |
},
|
| 523 |
{
|
| 524 |
-
"epoch":
|
| 525 |
-
"eval_accuracy": 0.
|
| 526 |
-
"eval_loss": 0.
|
| 527 |
-
"eval_runtime":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 528 |
"eval_samples_per_second": 49.665,
|
| 529 |
-
"eval_steps_per_second": 1.
|
| 530 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch":
|
| 534 |
-
"
|
| 535 |
-
"
|
| 536 |
-
"
|
| 537 |
-
"
|
| 538 |
-
"
|
| 539 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 540 |
}
|
| 541 |
],
|
| 542 |
-
"max_steps":
|
| 543 |
-
"num_train_epochs":
|
| 544 |
-
"total_flos":
|
| 545 |
"trial_name": null,
|
| 546 |
"trial_params": null
|
| 547 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.8832997987927566,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-agrivision/checkpoint-175",
|
| 4 |
+
"epoch": 50.0,
|
| 5 |
+
"global_step": 1750,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.29,
|
| 12 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 13 |
+
"loss": 1.2366,
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.57,
|
| 18 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 19 |
+
"loss": 1.1495,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
"step": 20
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.86,
|
| 24 |
+
"learning_rate": 8.571428571428573e-06,
|
| 25 |
+
"loss": 0.9718,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
"step": 30
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 1.0,
|
| 30 |
+
"eval_accuracy": 0.6398390342052314,
|
| 31 |
+
"eval_loss": 0.979684591293335,
|
| 32 |
+
"eval_runtime": 10.215,
|
| 33 |
+
"eval_samples_per_second": 48.654,
|
| 34 |
+
"eval_steps_per_second": 1.566,
|
| 35 |
+
"step": 35
|
| 36 |
},
|
| 37 |
{
|
| 38 |
+
"epoch": 1.14,
|
| 39 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 40 |
+
"loss": 0.7506,
|
| 41 |
+
"step": 40
|
|
|
|
|
|
|
|
|
|
| 42 |
},
|
| 43 |
{
|
| 44 |
+
"epoch": 1.43,
|
| 45 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 46 |
+
"loss": 0.5116,
|
| 47 |
"step": 50
|
| 48 |
},
|
| 49 |
{
|
| 50 |
+
"epoch": 1.71,
|
| 51 |
+
"learning_rate": 1.7142857142857145e-05,
|
| 52 |
+
"loss": 0.3792,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
"step": 60
|
| 54 |
},
|
| 55 |
{
|
| 56 |
+
"epoch": 2.0,
|
| 57 |
+
"learning_rate": 2e-05,
|
| 58 |
+
"loss": 0.2715,
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 2.0,
|
| 63 |
+
"eval_accuracy": 0.7484909456740443,
|
| 64 |
+
"eval_loss": 0.6352390646934509,
|
| 65 |
+
"eval_runtime": 10.1148,
|
| 66 |
+
"eval_samples_per_second": 49.136,
|
| 67 |
+
"eval_steps_per_second": 1.582,
|
| 68 |
"step": 70
|
| 69 |
},
|
| 70 |
{
|
| 71 |
+
"epoch": 2.29,
|
| 72 |
+
"learning_rate": 2.2857142857142858e-05,
|
| 73 |
+
"loss": 0.1984,
|
| 74 |
"step": 80
|
| 75 |
},
|
| 76 |
{
|
| 77 |
+
"epoch": 2.57,
|
| 78 |
+
"learning_rate": 2.5714285714285714e-05,
|
| 79 |
+
"loss": 0.1842,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
"step": 90
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"epoch": 2.86,
|
| 84 |
+
"learning_rate": 2.857142857142857e-05,
|
| 85 |
+
"loss": 0.1615,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 86 |
"step": 100
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 3.0,
|
| 90 |
+
"eval_accuracy": 0.8128772635814889,
|
| 91 |
+
"eval_loss": 0.4673824608325958,
|
| 92 |
+
"eval_runtime": 10.0382,
|
| 93 |
+
"eval_samples_per_second": 49.511,
|
| 94 |
+
"eval_steps_per_second": 1.594,
|
| 95 |
+
"step": 105
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"epoch": 3.14,
|
| 99 |
+
"learning_rate": 3.142857142857143e-05,
|
| 100 |
+
"loss": 0.1257,
|
| 101 |
+
"step": 110
|
|
|
|
|
|
|
|
|
|
| 102 |
},
|
| 103 |
{
|
| 104 |
+
"epoch": 3.43,
|
| 105 |
+
"learning_rate": 3.428571428571429e-05,
|
| 106 |
+
"loss": 0.103,
|
| 107 |
"step": 120
|
| 108 |
},
|
| 109 |
{
|
| 110 |
+
"epoch": 3.71,
|
| 111 |
+
"learning_rate": 3.7142857142857143e-05,
|
| 112 |
+
"loss": 0.1008,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 113 |
"step": 130
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"epoch": 4.0,
|
| 117 |
+
"learning_rate": 4e-05,
|
| 118 |
+
"loss": 0.0959,
|
| 119 |
"step": 140
|
| 120 |
},
|
| 121 |
{
|
| 122 |
+
"epoch": 4.0,
|
| 123 |
+
"eval_accuracy": 0.7847082494969819,
|
| 124 |
+
"eval_loss": 0.5569615960121155,
|
| 125 |
+
"eval_runtime": 10.1607,
|
| 126 |
+
"eval_samples_per_second": 48.914,
|
| 127 |
+
"eval_steps_per_second": 1.575,
|
| 128 |
"step": 140
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 4.29,
|
| 132 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 133 |
+
"loss": 0.0909,
|
| 134 |
"step": 150
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 4.57,
|
| 138 |
+
"learning_rate": 4.5714285714285716e-05,
|
| 139 |
+
"loss": 0.0678,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 140 |
"step": 160
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 4.86,
|
| 144 |
+
"learning_rate": 4.8571428571428576e-05,
|
| 145 |
+
"loss": 0.0563,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 146 |
"step": 170
|
| 147 |
},
|
| 148 |
{
|
| 149 |
+
"epoch": 5.0,
|
| 150 |
+
"eval_accuracy": 0.8832997987927566,
|
| 151 |
+
"eval_loss": 0.2902894914150238,
|
| 152 |
+
"eval_runtime": 10.0294,
|
| 153 |
+
"eval_samples_per_second": 49.554,
|
| 154 |
+
"eval_steps_per_second": 1.595,
|
| 155 |
+
"step": 175
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 5.14,
|
| 159 |
+
"learning_rate": 4.9841269841269845e-05,
|
| 160 |
+
"loss": 0.0465,
|
| 161 |
+
"step": 180
|
|
|
|
|
|
|
|
|
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 5.43,
|
| 165 |
+
"learning_rate": 4.9523809523809525e-05,
|
| 166 |
+
"loss": 0.0627,
|
| 167 |
"step": 190
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 5.71,
|
| 171 |
+
"learning_rate": 4.9206349206349204e-05,
|
| 172 |
+
"loss": 0.0262,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 173 |
"step": 200
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"epoch": 6.0,
|
| 177 |
+
"learning_rate": 4.888888888888889e-05,
|
| 178 |
+
"loss": 0.0824,
|
| 179 |
"step": 210
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 6.0,
|
| 183 |
+
"eval_accuracy": 0.8450704225352113,
|
| 184 |
+
"eval_loss": 0.41077324748039246,
|
| 185 |
+
"eval_runtime": 10.186,
|
| 186 |
+
"eval_samples_per_second": 48.793,
|
| 187 |
+
"eval_steps_per_second": 1.571,
|
| 188 |
"step": 210
|
| 189 |
},
|
| 190 |
{
|
| 191 |
+
"epoch": 6.29,
|
| 192 |
+
"learning_rate": 4.8571428571428576e-05,
|
| 193 |
+
"loss": 0.0562,
|
| 194 |
"step": 220
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 6.57,
|
| 198 |
+
"learning_rate": 4.8253968253968255e-05,
|
| 199 |
+
"loss": 0.0414,
|
| 200 |
+
"step": 230
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 6.86,
|
| 204 |
+
"learning_rate": 4.793650793650794e-05,
|
| 205 |
+
"loss": 0.0441,
|
| 206 |
+
"step": 240
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 7.0,
|
| 210 |
+
"eval_accuracy": 0.8209255533199196,
|
| 211 |
+
"eval_loss": 0.5062592029571533,
|
| 212 |
+
"eval_runtime": 10.1666,
|
| 213 |
+
"eval_samples_per_second": 48.886,
|
| 214 |
+
"eval_steps_per_second": 1.574,
|
| 215 |
+
"step": 245
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 7.14,
|
| 219 |
+
"learning_rate": 4.761904761904762e-05,
|
| 220 |
+
"loss": 0.0661,
|
| 221 |
+
"step": 250
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 7.43,
|
| 225 |
+
"learning_rate": 4.73015873015873e-05,
|
| 226 |
+
"loss": 0.0203,
|
| 227 |
+
"step": 260
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 7.71,
|
| 231 |
+
"learning_rate": 4.6984126984126986e-05,
|
| 232 |
+
"loss": 0.0342,
|
| 233 |
+
"step": 270
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 8.0,
|
| 237 |
+
"learning_rate": 4.666666666666667e-05,
|
| 238 |
+
"loss": 0.0306,
|
| 239 |
+
"step": 280
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 8.0,
|
| 243 |
+
"eval_accuracy": 0.8752515090543259,
|
| 244 |
+
"eval_loss": 0.30527463555336,
|
| 245 |
+
"eval_runtime": 10.0743,
|
| 246 |
+
"eval_samples_per_second": 49.333,
|
| 247 |
+
"eval_steps_per_second": 1.588,
|
| 248 |
+
"step": 280
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.29,
|
| 252 |
+
"learning_rate": 4.634920634920635e-05,
|
| 253 |
+
"loss": 0.0496,
|
| 254 |
+
"step": 290
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 8.57,
|
| 258 |
+
"learning_rate": 4.603174603174603e-05,
|
| 259 |
+
"loss": 0.0253,
|
| 260 |
+
"step": 300
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 8.86,
|
| 264 |
+
"learning_rate": 4.5714285714285716e-05,
|
| 265 |
+
"loss": 0.0279,
|
| 266 |
+
"step": 310
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 9.0,
|
| 270 |
+
"eval_accuracy": 0.8591549295774648,
|
| 271 |
+
"eval_loss": 0.44667041301727295,
|
| 272 |
+
"eval_runtime": 9.9268,
|
| 273 |
+
"eval_samples_per_second": 50.066,
|
| 274 |
+
"eval_steps_per_second": 1.612,
|
| 275 |
+
"step": 315
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 9.14,
|
| 279 |
+
"learning_rate": 4.5396825396825395e-05,
|
| 280 |
+
"loss": 0.0159,
|
| 281 |
+
"step": 320
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 9.43,
|
| 285 |
+
"learning_rate": 4.507936507936508e-05,
|
| 286 |
+
"loss": 0.0207,
|
| 287 |
+
"step": 330
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 9.71,
|
| 291 |
+
"learning_rate": 4.476190476190477e-05,
|
| 292 |
+
"loss": 0.0155,
|
| 293 |
+
"step": 340
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 10.0,
|
| 297 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 298 |
+
"loss": 0.0223,
|
| 299 |
+
"step": 350
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.0,
|
| 303 |
+
"eval_accuracy": 0.8551307847082495,
|
| 304 |
+
"eval_loss": 0.4208771884441376,
|
| 305 |
+
"eval_runtime": 10.2309,
|
| 306 |
+
"eval_samples_per_second": 48.578,
|
| 307 |
+
"eval_steps_per_second": 1.564,
|
| 308 |
+
"step": 350
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 10.29,
|
| 312 |
+
"learning_rate": 4.4126984126984126e-05,
|
| 313 |
+
"loss": 0.0152,
|
| 314 |
+
"step": 360
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 10.57,
|
| 318 |
+
"learning_rate": 4.380952380952381e-05,
|
| 319 |
+
"loss": 0.0172,
|
| 320 |
+
"step": 370
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 10.86,
|
| 324 |
+
"learning_rate": 4.34920634920635e-05,
|
| 325 |
+
"loss": 0.026,
|
| 326 |
+
"step": 380
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 11.0,
|
| 330 |
+
"eval_accuracy": 0.8531187122736419,
|
| 331 |
+
"eval_loss": 0.5075467228889465,
|
| 332 |
+
"eval_runtime": 10.086,
|
| 333 |
+
"eval_samples_per_second": 49.276,
|
| 334 |
+
"eval_steps_per_second": 1.586,
|
| 335 |
+
"step": 385
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 11.14,
|
| 339 |
+
"learning_rate": 4.317460317460318e-05,
|
| 340 |
+
"loss": 0.0204,
|
| 341 |
+
"step": 390
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 11.43,
|
| 345 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 346 |
+
"loss": 0.0267,
|
| 347 |
+
"step": 400
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 11.71,
|
| 351 |
+
"learning_rate": 4.253968253968254e-05,
|
| 352 |
+
"loss": 0.0227,
|
| 353 |
+
"step": 410
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 12.0,
|
| 357 |
+
"learning_rate": 4.222222222222222e-05,
|
| 358 |
+
"loss": 0.0344,
|
| 359 |
+
"step": 420
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 12.0,
|
| 363 |
+
"eval_accuracy": 0.8551307847082495,
|
| 364 |
+
"eval_loss": 0.4921689033508301,
|
| 365 |
+
"eval_runtime": 10.1843,
|
| 366 |
+
"eval_samples_per_second": 48.801,
|
| 367 |
+
"eval_steps_per_second": 1.571,
|
| 368 |
+
"step": 420
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 12.29,
|
| 372 |
+
"learning_rate": 4.190476190476191e-05,
|
| 373 |
+
"loss": 0.0175,
|
| 374 |
+
"step": 430
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 12.57,
|
| 378 |
+
"learning_rate": 4.1587301587301594e-05,
|
| 379 |
+
"loss": 0.0194,
|
| 380 |
+
"step": 440
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 12.86,
|
| 384 |
+
"learning_rate": 4.126984126984127e-05,
|
| 385 |
+
"loss": 0.0079,
|
| 386 |
+
"step": 450
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 13.0,
|
| 390 |
+
"eval_accuracy": 0.7806841046277666,
|
| 391 |
+
"eval_loss": 0.725607693195343,
|
| 392 |
+
"eval_runtime": 10.0703,
|
| 393 |
+
"eval_samples_per_second": 49.353,
|
| 394 |
+
"eval_steps_per_second": 1.589,
|
| 395 |
+
"step": 455
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 13.14,
|
| 399 |
+
"learning_rate": 4.095238095238095e-05,
|
| 400 |
+
"loss": 0.0169,
|
| 401 |
+
"step": 460
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 13.43,
|
| 405 |
+
"learning_rate": 4.063492063492064e-05,
|
| 406 |
+
"loss": 0.014,
|
| 407 |
+
"step": 470
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 13.71,
|
| 411 |
+
"learning_rate": 4.031746031746032e-05,
|
| 412 |
+
"loss": 0.0043,
|
| 413 |
+
"step": 480
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 14.0,
|
| 417 |
+
"learning_rate": 4e-05,
|
| 418 |
+
"loss": 0.0156,
|
| 419 |
+
"step": 490
|
| 420 |
},
|
| 421 |
{
|
| 422 |
+
"epoch": 14.0,
|
| 423 |
+
"eval_accuracy": 0.8370221327967807,
|
| 424 |
+
"eval_loss": 0.5176514387130737,
|
| 425 |
+
"eval_runtime": 9.8924,
|
| 426 |
+
"eval_samples_per_second": 50.241,
|
| 427 |
+
"eval_steps_per_second": 1.617,
|
| 428 |
+
"step": 490
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 14.29,
|
| 432 |
+
"learning_rate": 3.968253968253968e-05,
|
| 433 |
+
"loss": 0.0051,
|
| 434 |
+
"step": 500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 14.57,
|
| 438 |
+
"learning_rate": 3.936507936507937e-05,
|
| 439 |
+
"loss": 0.012,
|
| 440 |
+
"step": 510
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 14.86,
|
| 444 |
+
"learning_rate": 3.904761904761905e-05,
|
| 445 |
+
"loss": 0.0184,
|
| 446 |
+
"step": 520
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 15.0,
|
| 450 |
+
"eval_accuracy": 0.8350100603621731,
|
| 451 |
+
"eval_loss": 0.5857502222061157,
|
| 452 |
+
"eval_runtime": 10.2013,
|
| 453 |
+
"eval_samples_per_second": 48.719,
|
| 454 |
+
"eval_steps_per_second": 1.568,
|
| 455 |
+
"step": 525
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 15.14,
|
| 459 |
+
"learning_rate": 3.8730158730158734e-05,
|
| 460 |
+
"loss": 0.0045,
|
| 461 |
+
"step": 530
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 15.43,
|
| 465 |
+
"learning_rate": 3.841269841269842e-05,
|
| 466 |
"loss": 0.0052,
|
| 467 |
+
"step": 540
|
| 468 |
},
|
| 469 |
{
|
| 470 |
+
"epoch": 15.71,
|
| 471 |
+
"learning_rate": 3.809523809523809e-05,
|
| 472 |
+
"loss": 0.0074,
|
| 473 |
+
"step": 550
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 16.0,
|
| 477 |
+
"learning_rate": 3.777777777777778e-05,
|
| 478 |
+
"loss": 0.0121,
|
| 479 |
+
"step": 560
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 16.0,
|
| 483 |
+
"eval_accuracy": 0.8651911468812877,
|
| 484 |
+
"eval_loss": 0.4655379056930542,
|
| 485 |
+
"eval_runtime": 10.2086,
|
| 486 |
+
"eval_samples_per_second": 48.685,
|
| 487 |
+
"eval_steps_per_second": 1.567,
|
| 488 |
+
"step": 560
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 16.29,
|
| 492 |
+
"learning_rate": 3.7460317460317464e-05,
|
| 493 |
+
"loss": 0.0069,
|
| 494 |
+
"step": 570
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 16.57,
|
| 498 |
+
"learning_rate": 3.7142857142857143e-05,
|
| 499 |
+
"loss": 0.0215,
|
| 500 |
+
"step": 580
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 16.86,
|
| 504 |
+
"learning_rate": 3.682539682539683e-05,
|
| 505 |
+
"loss": 0.0089,
|
| 506 |
+
"step": 590
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 17.0,
|
| 510 |
+
"eval_accuracy": 0.7585513078470825,
|
| 511 |
+
"eval_loss": 0.9852096438407898,
|
| 512 |
+
"eval_runtime": 10.1571,
|
| 513 |
+
"eval_samples_per_second": 48.931,
|
| 514 |
+
"eval_steps_per_second": 1.575,
|
| 515 |
+
"step": 595
|
| 516 |
},
|
| 517 |
{
|
| 518 |
"epoch": 17.14,
|
| 519 |
+
"learning_rate": 3.650793650793651e-05,
|
| 520 |
+
"loss": 0.001,
|
| 521 |
+
"step": 600
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 17.43,
|
| 525 |
+
"learning_rate": 3.619047619047619e-05,
|
| 526 |
+
"loss": 0.0095,
|
| 527 |
+
"step": 610
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 17.71,
|
| 531 |
+
"learning_rate": 3.5873015873015874e-05,
|
| 532 |
+
"loss": 0.0028,
|
| 533 |
+
"step": 620
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 18.0,
|
| 537 |
+
"learning_rate": 3.555555555555556e-05,
|
| 538 |
+
"loss": 0.028,
|
| 539 |
+
"step": 630
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 18.0,
|
| 543 |
+
"eval_accuracy": 0.8571428571428571,
|
| 544 |
+
"eval_loss": 0.508708655834198,
|
| 545 |
+
"eval_runtime": 10.2049,
|
| 546 |
+
"eval_samples_per_second": 48.702,
|
| 547 |
+
"eval_steps_per_second": 1.568,
|
| 548 |
+
"step": 630
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 18.29,
|
| 552 |
+
"learning_rate": 3.523809523809524e-05,
|
| 553 |
+
"loss": 0.0024,
|
| 554 |
+
"step": 640
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 18.57,
|
| 558 |
+
"learning_rate": 3.492063492063492e-05,
|
| 559 |
+
"loss": 0.0017,
|
| 560 |
+
"step": 650
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 18.86,
|
| 564 |
+
"learning_rate": 3.4603174603174604e-05,
|
| 565 |
"loss": 0.003,
|
| 566 |
+
"step": 660
|
| 567 |
},
|
| 568 |
{
|
| 569 |
+
"epoch": 19.0,
|
| 570 |
+
"eval_accuracy": 0.8490945674044266,
|
| 571 |
+
"eval_loss": 0.544687807559967,
|
| 572 |
+
"eval_runtime": 9.9955,
|
| 573 |
+
"eval_samples_per_second": 49.722,
|
| 574 |
+
"eval_steps_per_second": 1.601,
|
| 575 |
+
"step": 665
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 19.14,
|
| 579 |
+
"learning_rate": 3.428571428571429e-05,
|
| 580 |
+
"loss": 0.0009,
|
| 581 |
+
"step": 670
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 19.43,
|
| 585 |
+
"learning_rate": 3.396825396825397e-05,
|
| 586 |
+
"loss": 0.0013,
|
| 587 |
+
"step": 680
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 19.71,
|
| 591 |
+
"learning_rate": 3.3650793650793656e-05,
|
| 592 |
+
"loss": 0.0011,
|
| 593 |
+
"step": 690
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 20.0,
|
| 597 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 598 |
+
"loss": 0.0015,
|
| 599 |
+
"step": 700
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 20.0,
|
| 603 |
+
"eval_accuracy": 0.8390342052313883,
|
| 604 |
+
"eval_loss": 0.5609642863273621,
|
| 605 |
+
"eval_runtime": 9.9297,
|
| 606 |
+
"eval_samples_per_second": 50.052,
|
| 607 |
+
"eval_steps_per_second": 1.611,
|
| 608 |
+
"step": 700
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 20.29,
|
| 612 |
+
"learning_rate": 3.3015873015873014e-05,
|
| 613 |
+
"loss": 0.0045,
|
| 614 |
+
"step": 710
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 20.57,
|
| 618 |
+
"learning_rate": 3.26984126984127e-05,
|
| 619 |
+
"loss": 0.0001,
|
| 620 |
+
"step": 720
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 20.86,
|
| 624 |
+
"learning_rate": 3.2380952380952386e-05,
|
| 625 |
+
"loss": 0.008,
|
| 626 |
+
"step": 730
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 21.0,
|
| 630 |
+
"eval_accuracy": 0.8571428571428571,
|
| 631 |
+
"eval_loss": 0.5701542496681213,
|
| 632 |
+
"eval_runtime": 10.1334,
|
| 633 |
+
"eval_samples_per_second": 49.046,
|
| 634 |
+
"eval_steps_per_second": 1.579,
|
| 635 |
+
"step": 735
|
| 636 |
},
|
| 637 |
{
|
| 638 |
+
"epoch": 21.14,
|
| 639 |
+
"learning_rate": 3.2063492063492065e-05,
|
| 640 |
+
"loss": 0.0007,
|
| 641 |
+
"step": 740
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 21.43,
|
| 645 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 646 |
+
"loss": 0.0013,
|
| 647 |
+
"step": 750
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 21.71,
|
| 651 |
+
"learning_rate": 3.142857142857143e-05,
|
| 652 |
+
"loss": 0.0008,
|
| 653 |
+
"step": 760
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"epoch": 22.0,
|
| 657 |
+
"learning_rate": 3.111111111111111e-05,
|
| 658 |
+
"loss": 0.0071,
|
| 659 |
+
"step": 770
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 22.0,
|
| 663 |
+
"eval_accuracy": 0.8450704225352113,
|
| 664 |
+
"eval_loss": 0.6042529344558716,
|
| 665 |
+
"eval_runtime": 10.0579,
|
| 666 |
+
"eval_samples_per_second": 49.414,
|
| 667 |
+
"eval_steps_per_second": 1.591,
|
| 668 |
+
"step": 770
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 22.29,
|
| 672 |
+
"learning_rate": 3.0793650793650796e-05,
|
| 673 |
+
"loss": 0.003,
|
| 674 |
+
"step": 780
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 22.57,
|
| 678 |
+
"learning_rate": 3.0476190476190482e-05,
|
| 679 |
+
"loss": 0.0027,
|
| 680 |
+
"step": 790
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 22.86,
|
| 684 |
+
"learning_rate": 3.0158730158730158e-05,
|
| 685 |
+
"loss": 0.0006,
|
| 686 |
+
"step": 800
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 23.0,
|
| 690 |
+
"eval_accuracy": 0.8450704225352113,
|
| 691 |
+
"eval_loss": 0.5951006412506104,
|
| 692 |
+
"eval_runtime": 10.1254,
|
| 693 |
+
"eval_samples_per_second": 49.085,
|
| 694 |
+
"eval_steps_per_second": 1.58,
|
| 695 |
+
"step": 805
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 23.14,
|
| 699 |
+
"learning_rate": 2.9841269841269844e-05,
|
| 700 |
+
"loss": 0.0092,
|
| 701 |
+
"step": 810
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 23.43,
|
| 705 |
+
"learning_rate": 2.9523809523809526e-05,
|
| 706 |
"loss": 0.0058,
|
| 707 |
+
"step": 820
|
| 708 |
},
|
| 709 |
{
|
| 710 |
+
"epoch": 23.71,
|
| 711 |
+
"learning_rate": 2.920634920634921e-05,
|
| 712 |
+
"loss": 0.0013,
|
| 713 |
+
"step": 830
|
|
|
|
|
|
|
|
|
|
| 714 |
},
|
| 715 |
{
|
| 716 |
+
"epoch": 24.0,
|
| 717 |
+
"learning_rate": 2.8888888888888888e-05,
|
| 718 |
+
"loss": 0.0077,
|
| 719 |
+
"step": 840
|
| 720 |
},
|
| 721 |
{
|
| 722 |
+
"epoch": 24.0,
|
| 723 |
+
"eval_accuracy": 0.8309859154929577,
|
| 724 |
+
"eval_loss": 0.6435739398002625,
|
| 725 |
+
"eval_runtime": 9.9406,
|
| 726 |
+
"eval_samples_per_second": 49.997,
|
| 727 |
+
"eval_steps_per_second": 1.61,
|
| 728 |
+
"step": 840
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 24.29,
|
| 732 |
+
"learning_rate": 2.857142857142857e-05,
|
| 733 |
+
"loss": 0.0025,
|
| 734 |
+
"step": 850
|
|
|
|
|
|
|
|
|
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 24.57,
|
| 738 |
+
"learning_rate": 2.8253968253968253e-05,
|
| 739 |
+
"loss": 0.0017,
|
| 740 |
+
"step": 860
|
| 741 |
},
|
| 742 |
{
|
| 743 |
+
"epoch": 24.86,
|
| 744 |
+
"learning_rate": 2.793650793650794e-05,
|
| 745 |
+
"loss": 0.0007,
|
| 746 |
+
"step": 870
|
|
|
|
|
|
|
|
|
|
| 747 |
},
|
| 748 |
{
|
| 749 |
+
"epoch": 25.0,
|
| 750 |
+
"eval_accuracy": 0.8792756539235412,
|
| 751 |
+
"eval_loss": 0.44388413429260254,
|
| 752 |
+
"eval_runtime": 10.0384,
|
| 753 |
+
"eval_samples_per_second": 49.51,
|
| 754 |
+
"eval_steps_per_second": 1.594,
|
| 755 |
+
"step": 875
|
| 756 |
},
|
| 757 |
{
|
| 758 |
+
"epoch": 25.14,
|
| 759 |
+
"learning_rate": 2.7619047619047622e-05,
|
| 760 |
+
"loss": 0.0016,
|
| 761 |
+
"step": 880
|
|
|
|
|
|
|
|
|
|
| 762 |
},
|
| 763 |
{
|
| 764 |
+
"epoch": 25.43,
|
| 765 |
+
"learning_rate": 2.73015873015873e-05,
|
| 766 |
+
"loss": 0.0066,
|
| 767 |
+
"step": 890
|
| 768 |
},
|
| 769 |
{
|
| 770 |
+
"epoch": 25.71,
|
| 771 |
+
"learning_rate": 2.6984126984126984e-05,
|
| 772 |
+
"loss": 0.0003,
|
| 773 |
+
"step": 900
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"epoch": 26.0,
|
| 777 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 778 |
+
"loss": 0.0081,
|
| 779 |
+
"step": 910
|
|
|
|
|
|
|
|
|
|
| 780 |
},
|
| 781 |
{
|
| 782 |
+
"epoch": 26.0,
|
| 783 |
+
"eval_accuracy": 0.8732394366197183,
|
| 784 |
+
"eval_loss": 0.46890053153038025,
|
| 785 |
+
"eval_runtime": 10.0258,
|
| 786 |
+
"eval_samples_per_second": 49.572,
|
| 787 |
+
"eval_steps_per_second": 1.596,
|
| 788 |
+
"step": 910
|
| 789 |
},
|
| 790 |
{
|
| 791 |
+
"epoch": 26.29,
|
| 792 |
+
"learning_rate": 2.6349206349206353e-05,
|
| 793 |
+
"loss": 0.0032,
|
| 794 |
+
"step": 920
|
|
|
|
|
|
|
|
|
|
| 795 |
},
|
| 796 |
{
|
| 797 |
+
"epoch": 26.57,
|
| 798 |
+
"learning_rate": 2.6031746031746035e-05,
|
| 799 |
+
"loss": 0.0019,
|
| 800 |
+
"step": 930
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 26.86,
|
| 804 |
+
"learning_rate": 2.5714285714285714e-05,
|
| 805 |
+
"loss": 0.0036,
|
| 806 |
+
"step": 940
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 27.0,
|
| 810 |
+
"eval_accuracy": 0.8712273641851107,
|
| 811 |
+
"eval_loss": 0.5057930946350098,
|
| 812 |
+
"eval_runtime": 9.9227,
|
| 813 |
+
"eval_samples_per_second": 50.087,
|
| 814 |
+
"eval_steps_per_second": 1.612,
|
| 815 |
+
"step": 945
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 27.14,
|
| 819 |
+
"learning_rate": 2.5396825396825397e-05,
|
| 820 |
+
"loss": 0.0037,
|
| 821 |
+
"step": 950
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 27.43,
|
| 825 |
+
"learning_rate": 2.507936507936508e-05,
|
| 826 |
"loss": 0.0008,
|
| 827 |
+
"step": 960
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 27.71,
|
| 831 |
+
"learning_rate": 2.4761904761904762e-05,
|
| 832 |
+
"loss": 0.0003,
|
| 833 |
+
"step": 970
|
|
|
|
|
|
|
|
|
|
| 834 |
},
|
| 835 |
{
|
| 836 |
+
"epoch": 28.0,
|
| 837 |
+
"learning_rate": 2.4444444444444445e-05,
|
| 838 |
+
"loss": 0.0004,
|
| 839 |
+
"step": 980
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 28.0,
|
| 843 |
+
"eval_accuracy": 0.8591549295774648,
|
| 844 |
+
"eval_loss": 0.5425885319709778,
|
| 845 |
+
"eval_runtime": 9.8796,
|
| 846 |
+
"eval_samples_per_second": 50.306,
|
| 847 |
+
"eval_steps_per_second": 1.62,
|
| 848 |
+
"step": 980
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 28.29,
|
| 852 |
+
"learning_rate": 2.4126984126984128e-05,
|
| 853 |
+
"loss": 0.0001,
|
| 854 |
+
"step": 990
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 28.57,
|
| 858 |
+
"learning_rate": 2.380952380952381e-05,
|
| 859 |
+
"loss": 0.0003,
|
| 860 |
+
"step": 1000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 28.86,
|
| 864 |
+
"learning_rate": 2.3492063492063493e-05,
|
| 865 |
+
"loss": 0.0007,
|
| 866 |
+
"step": 1010
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 29.0,
|
| 870 |
+
"eval_accuracy": 0.8631790744466801,
|
| 871 |
+
"eval_loss": 0.483518123626709,
|
| 872 |
+
"eval_runtime": 9.9988,
|
| 873 |
+
"eval_samples_per_second": 49.706,
|
| 874 |
+
"eval_steps_per_second": 1.6,
|
| 875 |
+
"step": 1015
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 29.14,
|
| 879 |
+
"learning_rate": 2.3174603174603175e-05,
|
| 880 |
+
"loss": 0.0006,
|
| 881 |
+
"step": 1020
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 29.43,
|
| 885 |
+
"learning_rate": 2.2857142857142858e-05,
|
| 886 |
+
"loss": 0.0008,
|
| 887 |
+
"step": 1030
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 29.71,
|
| 891 |
+
"learning_rate": 2.253968253968254e-05,
|
| 892 |
+
"loss": 0.005,
|
| 893 |
+
"step": 1040
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 30.0,
|
| 897 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 898 |
+
"loss": 0.005,
|
| 899 |
+
"step": 1050
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 30.0,
|
| 903 |
+
"eval_accuracy": 0.8651911468812877,
|
| 904 |
+
"eval_loss": 0.4958065152168274,
|
| 905 |
+
"eval_runtime": 10.0529,
|
| 906 |
+
"eval_samples_per_second": 49.438,
|
| 907 |
+
"eval_steps_per_second": 1.592,
|
| 908 |
+
"step": 1050
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 30.29,
|
| 912 |
+
"learning_rate": 2.1904761904761906e-05,
|
| 913 |
+
"loss": 0.0056,
|
| 914 |
+
"step": 1060
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 30.57,
|
| 918 |
+
"learning_rate": 2.158730158730159e-05,
|
| 919 |
+
"loss": 0.003,
|
| 920 |
+
"step": 1070
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 30.86,
|
| 924 |
+
"learning_rate": 2.126984126984127e-05,
|
| 925 |
+
"loss": 0.001,
|
| 926 |
+
"step": 1080
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 31.0,
|
| 930 |
+
"eval_accuracy": 0.8390342052313883,
|
| 931 |
+
"eval_loss": 0.7008472681045532,
|
| 932 |
+
"eval_runtime": 10.0335,
|
| 933 |
+
"eval_samples_per_second": 49.534,
|
| 934 |
+
"eval_steps_per_second": 1.595,
|
| 935 |
+
"step": 1085
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 31.14,
|
| 939 |
+
"learning_rate": 2.0952380952380954e-05,
|
| 940 |
+
"loss": 0.0007,
|
| 941 |
+
"step": 1090
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 31.43,
|
| 945 |
+
"learning_rate": 2.0634920634920636e-05,
|
| 946 |
+
"loss": 0.0005,
|
| 947 |
+
"step": 1100
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 31.71,
|
| 951 |
+
"learning_rate": 2.031746031746032e-05,
|
| 952 |
+
"loss": 0.0007,
|
| 953 |
+
"step": 1110
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 32.0,
|
| 957 |
+
"learning_rate": 2e-05,
|
| 958 |
+
"loss": 0.0022,
|
| 959 |
+
"step": 1120
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 32.0,
|
| 963 |
+
"eval_accuracy": 0.8631790744466801,
|
| 964 |
+
"eval_loss": 0.6209988594055176,
|
| 965 |
+
"eval_runtime": 9.9187,
|
| 966 |
+
"eval_samples_per_second": 50.107,
|
| 967 |
+
"eval_steps_per_second": 1.613,
|
| 968 |
+
"step": 1120
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 32.29,
|
| 972 |
+
"learning_rate": 1.9682539682539684e-05,
|
| 973 |
+
"loss": 0.0023,
|
| 974 |
+
"step": 1130
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 32.57,
|
| 978 |
+
"learning_rate": 1.9365079365079367e-05,
|
| 979 |
"loss": 0.0013,
|
| 980 |
+
"step": 1140
|
| 981 |
},
|
| 982 |
{
|
| 983 |
+
"epoch": 32.86,
|
| 984 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 985 |
+
"loss": 0.0029,
|
| 986 |
+
"step": 1150
|
|
|
|
|
|
|
|
|
|
| 987 |
},
|
| 988 |
{
|
| 989 |
+
"epoch": 33.0,
|
| 990 |
+
"eval_accuracy": 0.8470824949698189,
|
| 991 |
+
"eval_loss": 0.6328206062316895,
|
| 992 |
+
"eval_runtime": 9.9469,
|
| 993 |
+
"eval_samples_per_second": 49.966,
|
| 994 |
+
"eval_steps_per_second": 1.609,
|
| 995 |
+
"step": 1155
|
| 996 |
},
|
| 997 |
{
|
| 998 |
+
"epoch": 33.14,
|
| 999 |
+
"learning_rate": 1.8730158730158732e-05,
|
| 1000 |
+
"loss": 0.0007,
|
| 1001 |
+
"step": 1160
|
|
|
|
|
|
|
|
|
|
| 1002 |
},
|
| 1003 |
{
|
| 1004 |
+
"epoch": 33.43,
|
| 1005 |
+
"learning_rate": 1.8412698412698415e-05,
|
| 1006 |
+
"loss": 0.0121,
|
| 1007 |
+
"step": 1170
|
| 1008 |
},
|
| 1009 |
{
|
| 1010 |
+
"epoch": 33.71,
|
| 1011 |
+
"learning_rate": 1.8095238095238094e-05,
|
| 1012 |
+
"loss": 0.0006,
|
| 1013 |
+
"step": 1180
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 34.0,
|
| 1017 |
+
"learning_rate": 1.777777777777778e-05,
|
| 1018 |
+
"loss": 0.0001,
|
| 1019 |
+
"step": 1190
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 34.0,
|
| 1023 |
+
"eval_accuracy": 0.8551307847082495,
|
| 1024 |
+
"eval_loss": 0.5887029767036438,
|
| 1025 |
+
"eval_runtime": 10.1735,
|
| 1026 |
+
"eval_samples_per_second": 48.852,
|
| 1027 |
+
"eval_steps_per_second": 1.573,
|
| 1028 |
+
"step": 1190
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 34.29,
|
| 1032 |
+
"learning_rate": 1.746031746031746e-05,
|
| 1033 |
+
"loss": 0.0002,
|
| 1034 |
+
"step": 1200
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 34.57,
|
| 1038 |
+
"learning_rate": 1.7142857142857145e-05,
|
| 1039 |
+
"loss": 0.0001,
|
| 1040 |
+
"step": 1210
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 34.86,
|
| 1044 |
+
"learning_rate": 1.6825396825396828e-05,
|
| 1045 |
"loss": 0.0019,
|
| 1046 |
+
"step": 1220
|
| 1047 |
},
|
| 1048 |
{
|
| 1049 |
+
"epoch": 35.0,
|
| 1050 |
+
"eval_accuracy": 0.8390342052313883,
|
| 1051 |
+
"eval_loss": 0.6666443943977356,
|
| 1052 |
+
"eval_runtime": 9.9948,
|
| 1053 |
+
"eval_samples_per_second": 49.726,
|
| 1054 |
+
"eval_steps_per_second": 1.601,
|
| 1055 |
+
"step": 1225
|
| 1056 |
},
|
| 1057 |
{
|
| 1058 |
+
"epoch": 35.14,
|
| 1059 |
+
"learning_rate": 1.6507936507936507e-05,
|
| 1060 |
+
"loss": 0.0085,
|
| 1061 |
+
"step": 1230
|
| 1062 |
},
|
| 1063 |
{
|
| 1064 |
+
"epoch": 35.43,
|
| 1065 |
+
"learning_rate": 1.6190476190476193e-05,
|
| 1066 |
+
"loss": 0.0001,
|
| 1067 |
+
"step": 1240
|
|
|
|
|
|
|
|
|
|
| 1068 |
},
|
| 1069 |
{
|
| 1070 |
+
"epoch": 35.71,
|
| 1071 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 1072 |
+
"loss": 0.0008,
|
| 1073 |
+
"step": 1250
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 36.0,
|
| 1077 |
+
"learning_rate": 1.5555555555555555e-05,
|
| 1078 |
+
"loss": 0.0028,
|
| 1079 |
+
"step": 1260
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 36.0,
|
| 1083 |
+
"eval_accuracy": 0.8571428571428571,
|
| 1084 |
+
"eval_loss": 0.6371855735778809,
|
| 1085 |
+
"eval_runtime": 10.0695,
|
| 1086 |
+
"eval_samples_per_second": 49.357,
|
| 1087 |
+
"eval_steps_per_second": 1.589,
|
| 1088 |
+
"step": 1260
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 36.29,
|
| 1092 |
+
"learning_rate": 1.5238095238095241e-05,
|
| 1093 |
+
"loss": 0.0003,
|
| 1094 |
+
"step": 1270
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 36.57,
|
| 1098 |
+
"learning_rate": 1.4920634920634922e-05,
|
| 1099 |
+
"loss": 0.0001,
|
| 1100 |
+
"step": 1280
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 36.86,
|
| 1104 |
+
"learning_rate": 1.4603174603174605e-05,
|
| 1105 |
+
"loss": 0.0077,
|
| 1106 |
+
"step": 1290
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 37.0,
|
| 1110 |
+
"eval_accuracy": 0.8631790744466801,
|
| 1111 |
+
"eval_loss": 0.5971932411193848,
|
| 1112 |
+
"eval_runtime": 9.9137,
|
| 1113 |
+
"eval_samples_per_second": 50.132,
|
| 1114 |
+
"eval_steps_per_second": 1.614,
|
| 1115 |
+
"step": 1295
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 37.14,
|
| 1119 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 1120 |
+
"loss": 0.0002,
|
| 1121 |
+
"step": 1300
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 37.43,
|
| 1125 |
+
"learning_rate": 1.396825396825397e-05,
|
| 1126 |
+
"loss": 0.0007,
|
| 1127 |
+
"step": 1310
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 37.71,
|
| 1131 |
+
"learning_rate": 1.365079365079365e-05,
|
| 1132 |
"loss": 0.0006,
|
| 1133 |
+
"step": 1320
|
| 1134 |
},
|
| 1135 |
{
|
| 1136 |
+
"epoch": 38.0,
|
| 1137 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 1138 |
+
"loss": 0.0019,
|
| 1139 |
+
"step": 1330
|
| 1140 |
},
|
| 1141 |
{
|
| 1142 |
+
"epoch": 38.0,
|
| 1143 |
+
"eval_accuracy": 0.8752515090543259,
|
| 1144 |
+
"eval_loss": 0.5053229331970215,
|
| 1145 |
+
"eval_runtime": 10.0433,
|
| 1146 |
+
"eval_samples_per_second": 49.486,
|
| 1147 |
+
"eval_steps_per_second": 1.593,
|
| 1148 |
+
"step": 1330
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 38.29,
|
| 1152 |
+
"learning_rate": 1.3015873015873018e-05,
|
| 1153 |
+
"loss": 0.0024,
|
| 1154 |
+
"step": 1340
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 38.57,
|
| 1158 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 1159 |
+
"loss": 0.0019,
|
| 1160 |
+
"step": 1350
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 38.86,
|
| 1164 |
+
"learning_rate": 1.2380952380952381e-05,
|
| 1165 |
+
"loss": 0.0042,
|
| 1166 |
+
"step": 1360
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 39.0,
|
| 1170 |
+
"eval_accuracy": 0.8249496981891348,
|
| 1171 |
+
"eval_loss": 0.8244276642799377,
|
| 1172 |
+
"eval_runtime": 10.1573,
|
| 1173 |
+
"eval_samples_per_second": 48.93,
|
| 1174 |
+
"eval_steps_per_second": 1.575,
|
| 1175 |
+
"step": 1365
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 39.14,
|
| 1179 |
+
"learning_rate": 1.2063492063492064e-05,
|
| 1180 |
+
"loss": 0.0003,
|
| 1181 |
+
"step": 1370
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 39.43,
|
| 1185 |
+
"learning_rate": 1.1746031746031746e-05,
|
| 1186 |
+
"loss": 0.004,
|
| 1187 |
+
"step": 1380
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 39.71,
|
| 1191 |
+
"learning_rate": 1.1428571428571429e-05,
|
| 1192 |
+
"loss": 0.0002,
|
| 1193 |
+
"step": 1390
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 40.0,
|
| 1197 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 1198 |
+
"loss": 0.0022,
|
| 1199 |
+
"step": 1400
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 40.0,
|
| 1203 |
+
"eval_accuracy": 0.8410462776659959,
|
| 1204 |
+
"eval_loss": 0.7625672221183777,
|
| 1205 |
+
"eval_runtime": 10.1275,
|
| 1206 |
+
"eval_samples_per_second": 49.074,
|
| 1207 |
+
"eval_steps_per_second": 1.58,
|
| 1208 |
+
"step": 1400
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 40.29,
|
| 1212 |
+
"learning_rate": 1.0793650793650794e-05,
|
| 1213 |
+
"loss": 0.0019,
|
| 1214 |
+
"step": 1410
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 40.57,
|
| 1218 |
+
"learning_rate": 1.0476190476190477e-05,
|
| 1219 |
+
"loss": 0.0001,
|
| 1220 |
+
"step": 1420
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 40.86,
|
| 1224 |
+
"learning_rate": 1.015873015873016e-05,
|
| 1225 |
+
"loss": 0.0036,
|
| 1226 |
+
"step": 1430
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 41.0,
|
| 1230 |
+
"eval_accuracy": 0.8410462776659959,
|
| 1231 |
+
"eval_loss": 0.6884453296661377,
|
| 1232 |
+
"eval_runtime": 9.9577,
|
| 1233 |
+
"eval_samples_per_second": 49.911,
|
| 1234 |
+
"eval_steps_per_second": 1.607,
|
| 1235 |
+
"step": 1435
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 41.14,
|
| 1239 |
+
"learning_rate": 9.841269841269842e-06,
|
| 1240 |
+
"loss": 0.0001,
|
| 1241 |
+
"step": 1440
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 41.43,
|
| 1245 |
+
"learning_rate": 9.523809523809523e-06,
|
| 1246 |
+
"loss": 0.0004,
|
| 1247 |
+
"step": 1450
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 41.71,
|
| 1251 |
+
"learning_rate": 9.206349206349207e-06,
|
| 1252 |
+
"loss": 0.002,
|
| 1253 |
+
"step": 1460
|
| 1254 |
+
},
|
| 1255 |
+
{
|
| 1256 |
+
"epoch": 42.0,
|
| 1257 |
+
"learning_rate": 8.88888888888889e-06,
|
| 1258 |
+
"loss": 0.0016,
|
| 1259 |
+
"step": 1470
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 42.0,
|
| 1263 |
+
"eval_accuracy": 0.8410462776659959,
|
| 1264 |
+
"eval_loss": 0.6704004406929016,
|
| 1265 |
+
"eval_runtime": 10.0751,
|
| 1266 |
+
"eval_samples_per_second": 49.33,
|
| 1267 |
+
"eval_steps_per_second": 1.588,
|
| 1268 |
+
"step": 1470
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 42.29,
|
| 1272 |
+
"learning_rate": 8.571428571428573e-06,
|
| 1273 |
+
"loss": 0.0003,
|
| 1274 |
+
"step": 1480
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 42.57,
|
| 1278 |
+
"learning_rate": 8.253968253968254e-06,
|
| 1279 |
+
"loss": 0.0003,
|
| 1280 |
+
"step": 1490
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 42.86,
|
| 1284 |
+
"learning_rate": 7.936507936507936e-06,
|
| 1285 |
+
"loss": 0.0011,
|
| 1286 |
+
"step": 1500
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 43.0,
|
| 1290 |
+
"eval_accuracy": 0.8531187122736419,
|
| 1291 |
+
"eval_loss": 0.5821227431297302,
|
| 1292 |
+
"eval_runtime": 9.9252,
|
| 1293 |
+
"eval_samples_per_second": 50.074,
|
| 1294 |
+
"eval_steps_per_second": 1.612,
|
| 1295 |
+
"step": 1505
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 43.14,
|
| 1299 |
+
"learning_rate": 7.6190476190476205e-06,
|
| 1300 |
+
"loss": 0.0011,
|
| 1301 |
+
"step": 1510
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 43.43,
|
| 1305 |
+
"learning_rate": 7.301587301587302e-06,
|
| 1306 |
+
"loss": 0.0002,
|
| 1307 |
+
"step": 1520
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 43.71,
|
| 1311 |
+
"learning_rate": 6.984126984126985e-06,
|
| 1312 |
+
"loss": 0.0,
|
| 1313 |
+
"step": 1530
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 44.0,
|
| 1317 |
+
"learning_rate": 6.666666666666667e-06,
|
| 1318 |
+
"loss": 0.0001,
|
| 1319 |
+
"step": 1540
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 44.0,
|
| 1323 |
+
"eval_accuracy": 0.8571428571428571,
|
| 1324 |
+
"eval_loss": 0.5814576148986816,
|
| 1325 |
+
"eval_runtime": 10.0565,
|
| 1326 |
+
"eval_samples_per_second": 49.421,
|
| 1327 |
+
"eval_steps_per_second": 1.591,
|
| 1328 |
+
"step": 1540
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 44.29,
|
| 1332 |
+
"learning_rate": 6.349206349206349e-06,
|
| 1333 |
+
"loss": 0.0012,
|
| 1334 |
+
"step": 1550
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 44.57,
|
| 1338 |
+
"learning_rate": 6.031746031746032e-06,
|
| 1339 |
+
"loss": 0.0006,
|
| 1340 |
+
"step": 1560
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 44.86,
|
| 1344 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 1345 |
+
"loss": 0.0003,
|
| 1346 |
+
"step": 1570
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 45.0,
|
| 1350 |
+
"eval_accuracy": 0.8430583501006036,
|
| 1351 |
+
"eval_loss": 0.6694456934928894,
|
| 1352 |
+
"eval_runtime": 10.1535,
|
| 1353 |
+
"eval_samples_per_second": 48.949,
|
| 1354 |
+
"eval_steps_per_second": 1.576,
|
| 1355 |
+
"step": 1575
|
| 1356 |
+
},
|
| 1357 |
+
{
|
| 1358 |
+
"epoch": 45.14,
|
| 1359 |
+
"learning_rate": 5.396825396825397e-06,
|
| 1360 |
+
"loss": 0.0001,
|
| 1361 |
+
"step": 1580
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 45.43,
|
| 1365 |
+
"learning_rate": 5.07936507936508e-06,
|
| 1366 |
+
"loss": 0.0002,
|
| 1367 |
+
"step": 1590
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 45.71,
|
| 1371 |
+
"learning_rate": 4.7619047619047615e-06,
|
| 1372 |
+
"loss": 0.0001,
|
| 1373 |
+
"step": 1600
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 46.0,
|
| 1377 |
+
"learning_rate": 4.444444444444445e-06,
|
| 1378 |
+
"loss": 0.0007,
|
| 1379 |
+
"step": 1610
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 46.0,
|
| 1383 |
+
"eval_accuracy": 0.8430583501006036,
|
| 1384 |
+
"eval_loss": 0.687676727771759,
|
| 1385 |
+
"eval_runtime": 10.0071,
|
| 1386 |
"eval_samples_per_second": 49.665,
|
| 1387 |
+
"eval_steps_per_second": 1.599,
|
| 1388 |
+
"step": 1610
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 46.29,
|
| 1392 |
+
"learning_rate": 4.126984126984127e-06,
|
| 1393 |
+
"loss": 0.0024,
|
| 1394 |
+
"step": 1620
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 46.57,
|
| 1398 |
+
"learning_rate": 3.8095238095238102e-06,
|
| 1399 |
+
"loss": 0.0001,
|
| 1400 |
+
"step": 1630
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 46.86,
|
| 1404 |
+
"learning_rate": 3.4920634920634924e-06,
|
| 1405 |
+
"loss": 0.0,
|
| 1406 |
+
"step": 1640
|
| 1407 |
},
|
| 1408 |
{
|
| 1409 |
+
"epoch": 47.0,
|
| 1410 |
+
"eval_accuracy": 0.8390342052313883,
|
| 1411 |
+
"eval_loss": 0.6862754821777344,
|
| 1412 |
+
"eval_runtime": 10.0393,
|
| 1413 |
+
"eval_samples_per_second": 49.505,
|
| 1414 |
+
"eval_steps_per_second": 1.594,
|
| 1415 |
+
"step": 1645
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 47.14,
|
| 1419 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 1420 |
+
"loss": 0.0,
|
| 1421 |
+
"step": 1650
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 47.43,
|
| 1425 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 1426 |
+
"loss": 0.0001,
|
| 1427 |
+
"step": 1660
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 47.71,
|
| 1431 |
+
"learning_rate": 2.53968253968254e-06,
|
| 1432 |
+
"loss": 0.0,
|
| 1433 |
+
"step": 1670
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 48.0,
|
| 1437 |
+
"learning_rate": 2.2222222222222225e-06,
|
| 1438 |
+
"loss": 0.0,
|
| 1439 |
+
"step": 1680
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 48.0,
|
| 1443 |
+
"eval_accuracy": 0.8430583501006036,
|
| 1444 |
+
"eval_loss": 0.6967443823814392,
|
| 1445 |
+
"eval_runtime": 9.9265,
|
| 1446 |
+
"eval_samples_per_second": 50.068,
|
| 1447 |
+
"eval_steps_per_second": 1.612,
|
| 1448 |
+
"step": 1680
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 48.29,
|
| 1452 |
+
"learning_rate": 1.9047619047619051e-06,
|
| 1453 |
+
"loss": 0.0003,
|
| 1454 |
+
"step": 1690
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 48.57,
|
| 1458 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 1459 |
+
"loss": 0.0001,
|
| 1460 |
+
"step": 1700
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 48.86,
|
| 1464 |
+
"learning_rate": 1.26984126984127e-06,
|
| 1465 |
+
"loss": 0.0001,
|
| 1466 |
+
"step": 1710
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 49.0,
|
| 1470 |
+
"eval_accuracy": 0.8410462776659959,
|
| 1471 |
+
"eval_loss": 0.6850759387016296,
|
| 1472 |
+
"eval_runtime": 10.186,
|
| 1473 |
+
"eval_samples_per_second": 48.792,
|
| 1474 |
+
"eval_steps_per_second": 1.571,
|
| 1475 |
+
"step": 1715
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 49.14,
|
| 1479 |
+
"learning_rate": 9.523809523809526e-07,
|
| 1480 |
+
"loss": 0.0005,
|
| 1481 |
+
"step": 1720
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 49.43,
|
| 1485 |
+
"learning_rate": 6.34920634920635e-07,
|
| 1486 |
+
"loss": 0.0001,
|
| 1487 |
+
"step": 1730
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 49.71,
|
| 1491 |
+
"learning_rate": 3.174603174603175e-07,
|
| 1492 |
+
"loss": 0.0001,
|
| 1493 |
+
"step": 1740
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 50.0,
|
| 1497 |
+
"learning_rate": 0.0,
|
| 1498 |
+
"loss": 0.0,
|
| 1499 |
+
"step": 1750
|
| 1500 |
+
},
|
| 1501 |
+
{
|
| 1502 |
+
"epoch": 50.0,
|
| 1503 |
+
"eval_accuracy": 0.8410462776659959,
|
| 1504 |
+
"eval_loss": 0.6828312277793884,
|
| 1505 |
+
"eval_runtime": 9.963,
|
| 1506 |
+
"eval_samples_per_second": 49.885,
|
| 1507 |
+
"eval_steps_per_second": 1.606,
|
| 1508 |
+
"step": 1750
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 50.0,
|
| 1512 |
+
"step": 1750,
|
| 1513 |
+
"total_flos": 5.554381127113728e+18,
|
| 1514 |
+
"train_loss": 0.044299079362735416,
|
| 1515 |
+
"train_runtime": 9406.6232,
|
| 1516 |
+
"train_samples_per_second": 23.755,
|
| 1517 |
+
"train_steps_per_second": 0.186
|
| 1518 |
}
|
| 1519 |
],
|
| 1520 |
+
"max_steps": 1750,
|
| 1521 |
+
"num_train_epochs": 50,
|
| 1522 |
+
"total_flos": 5.554381127113728e+18,
|
| 1523 |
"trial_name": null,
|
| 1524 |
"trial_params": null
|
| 1525 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3439
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4c59fb9d277b60c557b7f361b4f4771289f10e485ce9b8b242f8ca8c10071067
|
| 3 |
size 3439
|