Commit
•
3e8a24b
1
Parent(s):
125964e
End of training
Browse files
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_accuracy": 0.8795454545454545,
|
| 4 |
"eval_loss": 0.3292839825153351,
|
| 5 |
"eval_runtime": 11.0507,
|
| 6 |
"eval_samples_per_second": 79.633,
|
| 7 |
"eval_steps_per_second": 4.977,
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss": 0.
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
"eval_accuracy": 0.8795454545454545,
|
| 4 |
"eval_loss": 0.3292839825153351,
|
| 5 |
"eval_runtime": 11.0507,
|
| 6 |
"eval_samples_per_second": 79.633,
|
| 7 |
"eval_steps_per_second": 4.977,
|
| 8 |
+
"total_flos": 8.960328656655483e+18,
|
| 9 |
+
"train_loss": 0.807003553578111,
|
| 10 |
+
"train_runtime": 6787.6894,
|
| 11 |
+
"train_samples_per_second": 5.732,
|
| 12 |
+
"train_steps_per_second": 0.09
|
| 13 |
}
|
runs/Jul05_16-55-02_c8d5d7484861/events.out.tfevents.1688583640.c8d5d7484861.2170.7
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b8d568d9f820c093e6bbb55d7022f1cd7e089471633a582df87bcb6af8832e3
|
| 3 |
+
size 456
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss": 0.
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 8.960328656655483e+18,
|
| 4 |
+
"train_loss": 0.807003553578111,
|
| 5 |
+
"train_runtime": 6787.6894,
|
| 6 |
+
"train_samples_per_second": 5.732,
|
| 7 |
+
"train_steps_per_second": 0.09
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,277 +1,481 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-eurosat-50/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":
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
-
"epoch": 0.
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss":
|
| 20 |
"step": 20
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss":
|
| 26 |
"step": 30
|
| 27 |
},
|
| 28 |
{
|
| 29 |
-
"epoch": 0.
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss":
|
| 32 |
"step": 40
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
-
"learning_rate":
|
| 37 |
-
"loss":
|
| 38 |
"step": 50
|
| 39 |
},
|
| 40 |
{
|
| 41 |
-
"epoch":
|
| 42 |
-
"
|
| 43 |
-
"
|
| 44 |
-
"
|
| 45 |
-
"eval_samples_per_second": 80.581,
|
| 46 |
-
"eval_steps_per_second": 5.036,
|
| 47 |
-
"step": 55
|
| 48 |
},
|
| 49 |
{
|
| 50 |
-
"epoch": 1.
|
| 51 |
-
"
|
| 52 |
-
"
|
| 53 |
-
"
|
|
|
|
|
|
|
|
|
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"epoch": 1.
|
| 57 |
-
"learning_rate":
|
| 58 |
-
"loss":
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
-
"epoch": 1.
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss":
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch": 1.
|
| 69 |
-
"learning_rate":
|
| 70 |
-
"loss":
|
| 71 |
"step": 90
|
| 72 |
},
|
| 73 |
{
|
| 74 |
-
"epoch": 1.
|
| 75 |
-
"learning_rate":
|
| 76 |
-
"loss":
|
| 77 |
"step": 100
|
| 78 |
},
|
| 79 |
{
|
| 80 |
-
"epoch":
|
| 81 |
-
"learning_rate":
|
| 82 |
-
"loss":
|
| 83 |
"step": 110
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"epoch":
|
| 87 |
-
"
|
| 88 |
-
"
|
| 89 |
-
"
|
| 90 |
-
"eval_samples_per_second": 80.65,
|
| 91 |
-
"eval_steps_per_second": 5.041,
|
| 92 |
-
"step": 110
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch": 2.
|
| 96 |
-
"
|
| 97 |
-
"
|
| 98 |
-
"
|
|
|
|
|
|
|
|
|
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"epoch": 2.
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss":
|
| 104 |
"step": 130
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"epoch": 2.
|
| 108 |
-
"learning_rate":
|
| 109 |
-
"loss":
|
| 110 |
"step": 140
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"epoch": 2.
|
| 114 |
-
"learning_rate":
|
| 115 |
-
"loss":
|
| 116 |
"step": 150
|
| 117 |
},
|
| 118 |
{
|
| 119 |
-
"epoch": 2.
|
| 120 |
-
"learning_rate":
|
| 121 |
-
"loss":
|
| 122 |
"step": 160
|
| 123 |
},
|
| 124 |
{
|
| 125 |
-
"epoch":
|
| 126 |
-
"
|
| 127 |
-
"
|
| 128 |
-
"eval_runtime": 11.069,
|
| 129 |
-
"eval_samples_per_second": 79.501,
|
| 130 |
-
"eval_steps_per_second": 4.969,
|
| 131 |
-
"step": 165
|
| 132 |
-
},
|
| 133 |
-
{
|
| 134 |
-
"epoch": 3.09,
|
| 135 |
-
"learning_rate": 9.696969696969698e-06,
|
| 136 |
-
"loss": 0.3096,
|
| 137 |
"step": 170
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch":
|
| 141 |
-
"learning_rate":
|
| 142 |
-
"loss": 0.
|
| 143 |
"step": 180
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"epoch": 3.
|
| 147 |
-
"
|
| 148 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 149 |
"step": 190
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 3.
|
| 153 |
-
"learning_rate":
|
| 154 |
-
"loss": 0.
|
| 155 |
"step": 200
|
| 156 |
},
|
| 157 |
{
|
| 158 |
-
"epoch": 3.
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss": 0.
|
| 161 |
"step": 210
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch":
|
| 165 |
-
"learning_rate":
|
| 166 |
-
"loss": 0.
|
| 167 |
-
"step": 220
|
| 168 |
-
},
|
| 169 |
-
{
|
| 170 |
-
"epoch": 4.0,
|
| 171 |
-
"eval_accuracy": 0.9397727272727273,
|
| 172 |
-
"eval_loss": 0.19657106697559357,
|
| 173 |
-
"eval_runtime": 10.9849,
|
| 174 |
-
"eval_samples_per_second": 80.11,
|
| 175 |
-
"eval_steps_per_second": 5.007,
|
| 176 |
"step": 220
|
| 177 |
},
|
| 178 |
{
|
| 179 |
-
"epoch":
|
| 180 |
-
"learning_rate":
|
| 181 |
-
"loss": 0.
|
| 182 |
"step": 230
|
| 183 |
},
|
| 184 |
{
|
| 185 |
-
"epoch":
|
| 186 |
-
"learning_rate":
|
| 187 |
-
"loss": 0.
|
| 188 |
"step": 240
|
| 189 |
},
|
| 190 |
{
|
| 191 |
-
"epoch": 4.
|
| 192 |
-
"
|
| 193 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 194 |
"step": 250
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch": 4.
|
| 198 |
-
"learning_rate":
|
| 199 |
-
"loss": 0.
|
| 200 |
"step": 260
|
| 201 |
},
|
| 202 |
{
|
| 203 |
-
"epoch": 4.
|
| 204 |
-
"learning_rate":
|
| 205 |
-
"loss": 0.
|
| 206 |
"step": 270
|
| 207 |
},
|
| 208 |
{
|
| 209 |
-
"epoch":
|
| 210 |
-
"
|
| 211 |
-
"
|
| 212 |
-
"eval_runtime": 10.9353,
|
| 213 |
-
"eval_samples_per_second": 80.474,
|
| 214 |
-
"eval_steps_per_second": 5.03,
|
| 215 |
-
"step": 275
|
| 216 |
-
},
|
| 217 |
-
{
|
| 218 |
-
"epoch": 5.09,
|
| 219 |
-
"learning_rate": 3.0303030303030305e-06,
|
| 220 |
-
"loss": 0.219,
|
| 221 |
"step": 280
|
| 222 |
},
|
| 223 |
{
|
| 224 |
-
"epoch":
|
| 225 |
-
"learning_rate":
|
| 226 |
-
"loss": 0.
|
| 227 |
"step": 290
|
| 228 |
},
|
| 229 |
{
|
| 230 |
-
"epoch":
|
| 231 |
-
"learning_rate":
|
| 232 |
-
"loss": 0.
|
| 233 |
"step": 300
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"epoch": 5.
|
| 237 |
-
"
|
| 238 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 239 |
"step": 310
|
| 240 |
},
|
| 241 |
{
|
| 242 |
-
"epoch": 5.
|
| 243 |
-
"learning_rate":
|
| 244 |
-
"loss": 0.
|
| 245 |
"step": 320
|
| 246 |
},
|
| 247 |
{
|
| 248 |
-
"epoch":
|
| 249 |
-
"learning_rate":
|
| 250 |
-
"loss": 0.
|
| 251 |
"step": 330
|
| 252 |
},
|
| 253 |
{
|
| 254 |
-
"epoch":
|
| 255 |
-
"
|
| 256 |
-
"
|
| 257 |
-
"
|
| 258 |
-
|
| 259 |
-
|
| 260 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 261 |
},
|
| 262 |
{
|
| 263 |
"epoch": 6.0,
|
| 264 |
-
"
|
| 265 |
-
"
|
| 266 |
-
"
|
| 267 |
-
"
|
| 268 |
-
"
|
| 269 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 270 |
}
|
| 271 |
],
|
| 272 |
-
"max_steps":
|
| 273 |
-
"num_train_epochs":
|
| 274 |
-
"total_flos":
|
| 275 |
"trial_name": null,
|
| 276 |
"trial_params": null
|
| 277 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.9732785200411099,
|
| 3 |
+
"best_model_checkpoint": "swinv2-large-patch4-window12to16-192to256-22kto1k-ft-finetuned-eurosat-50/checkpoint-549",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 610,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.16,
|
| 12 |
+
"learning_rate": 3.278688524590164e-07,
|
| 13 |
+
"loss": 2.1255,
|
| 14 |
"step": 10
|
| 15 |
},
|
| 16 |
{
|
| 17 |
+
"epoch": 0.33,
|
| 18 |
+
"learning_rate": 6.557377049180328e-07,
|
| 19 |
+
"loss": 2.045,
|
| 20 |
"step": 20
|
| 21 |
},
|
| 22 |
{
|
| 23 |
+
"epoch": 0.49,
|
| 24 |
+
"learning_rate": 9.836065573770493e-07,
|
| 25 |
+
"loss": 2.0287,
|
| 26 |
"step": 30
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.66,
|
| 30 |
+
"learning_rate": 1.3114754098360657e-06,
|
| 31 |
+
"loss": 2.0054,
|
| 32 |
"step": 40
|
| 33 |
},
|
| 34 |
{
|
| 35 |
+
"epoch": 0.82,
|
| 36 |
+
"learning_rate": 1.6393442622950819e-06,
|
| 37 |
+
"loss": 1.9418,
|
| 38 |
"step": 50
|
| 39 |
},
|
| 40 |
{
|
| 41 |
+
"epoch": 0.98,
|
| 42 |
+
"learning_rate": 1.9672131147540985e-06,
|
| 43 |
+
"loss": 1.9035,
|
| 44 |
+
"step": 60
|
|
|
|
|
|
|
|
|
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 1.0,
|
| 48 |
+
"eval_accuracy": 0.27132579650565264,
|
| 49 |
+
"eval_loss": 1.8945504426956177,
|
| 50 |
+
"eval_runtime": 12.371,
|
| 51 |
+
"eval_samples_per_second": 78.652,
|
| 52 |
+
"eval_steps_per_second": 4.931,
|
| 53 |
+
"step": 61
|
| 54 |
},
|
| 55 |
{
|
| 56 |
+
"epoch": 1.15,
|
| 57 |
+
"learning_rate": 2.295081967213115e-06,
|
| 58 |
+
"loss": 1.8717,
|
| 59 |
"step": 70
|
| 60 |
},
|
| 61 |
{
|
| 62 |
+
"epoch": 1.31,
|
| 63 |
+
"learning_rate": 2.6229508196721314e-06,
|
| 64 |
+
"loss": 1.8202,
|
| 65 |
"step": 80
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 1.48,
|
| 69 |
+
"learning_rate": 2.9508196721311478e-06,
|
| 70 |
+
"loss": 1.7631,
|
| 71 |
"step": 90
|
| 72 |
},
|
| 73 |
{
|
| 74 |
+
"epoch": 1.64,
|
| 75 |
+
"learning_rate": 3.2786885245901638e-06,
|
| 76 |
+
"loss": 1.7051,
|
| 77 |
"step": 100
|
| 78 |
},
|
| 79 |
{
|
| 80 |
+
"epoch": 1.8,
|
| 81 |
+
"learning_rate": 3.6065573770491806e-06,
|
| 82 |
+
"loss": 1.6242,
|
| 83 |
"step": 110
|
| 84 |
},
|
| 85 |
{
|
| 86 |
+
"epoch": 1.97,
|
| 87 |
+
"learning_rate": 3.934426229508197e-06,
|
| 88 |
+
"loss": 1.4731,
|
| 89 |
+
"step": 120
|
|
|
|
|
|
|
|
|
|
| 90 |
},
|
| 91 |
{
|
| 92 |
+
"epoch": 2.0,
|
| 93 |
+
"eval_accuracy": 0.5560123329907503,
|
| 94 |
+
"eval_loss": 1.2931360006332397,
|
| 95 |
+
"eval_runtime": 12.3906,
|
| 96 |
+
"eval_samples_per_second": 78.527,
|
| 97 |
+
"eval_steps_per_second": 4.923,
|
| 98 |
+
"step": 122
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"epoch": 2.13,
|
| 102 |
+
"learning_rate": 4.2622950819672135e-06,
|
| 103 |
+
"loss": 1.3791,
|
| 104 |
"step": 130
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 2.3,
|
| 108 |
+
"learning_rate": 4.59016393442623e-06,
|
| 109 |
+
"loss": 1.2578,
|
| 110 |
"step": 140
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"epoch": 2.46,
|
| 114 |
+
"learning_rate": 4.918032786885246e-06,
|
| 115 |
+
"loss": 1.1893,
|
| 116 |
"step": 150
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"epoch": 2.62,
|
| 120 |
+
"learning_rate": 5.245901639344263e-06,
|
| 121 |
+
"loss": 1.037,
|
| 122 |
"step": 160
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 2.79,
|
| 126 |
+
"learning_rate": 5.573770491803278e-06,
|
| 127 |
+
"loss": 1.0718,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 128 |
"step": 170
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"epoch": 2.95,
|
| 132 |
+
"learning_rate": 5.9016393442622956e-06,
|
| 133 |
+
"loss": 0.9549,
|
| 134 |
"step": 180
|
| 135 |
},
|
| 136 |
{
|
| 137 |
+
"epoch": 3.0,
|
| 138 |
+
"eval_accuracy": 0.6998972250770812,
|
| 139 |
+
"eval_loss": 0.7530122399330139,
|
| 140 |
+
"eval_runtime": 12.2864,
|
| 141 |
+
"eval_samples_per_second": 79.193,
|
| 142 |
+
"eval_steps_per_second": 4.965,
|
| 143 |
+
"step": 183
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 3.11,
|
| 147 |
+
"learning_rate": 6.229508196721312e-06,
|
| 148 |
+
"loss": 1.0102,
|
| 149 |
"step": 190
|
| 150 |
},
|
| 151 |
{
|
| 152 |
+
"epoch": 3.28,
|
| 153 |
+
"learning_rate": 6.5573770491803276e-06,
|
| 154 |
+
"loss": 0.8888,
|
| 155 |
"step": 200
|
| 156 |
},
|
| 157 |
{
|
| 158 |
+
"epoch": 3.44,
|
| 159 |
+
"learning_rate": 6.885245901639345e-06,
|
| 160 |
+
"loss": 0.8389,
|
| 161 |
"step": 210
|
| 162 |
},
|
| 163 |
{
|
| 164 |
+
"epoch": 3.61,
|
| 165 |
+
"learning_rate": 7.213114754098361e-06,
|
| 166 |
+
"loss": 0.8072,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 167 |
"step": 220
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 3.77,
|
| 171 |
+
"learning_rate": 7.540983606557377e-06,
|
| 172 |
+
"loss": 0.748,
|
| 173 |
"step": 230
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"epoch": 3.93,
|
| 177 |
+
"learning_rate": 7.868852459016394e-06,
|
| 178 |
+
"loss": 0.7375,
|
| 179 |
"step": 240
|
| 180 |
},
|
| 181 |
{
|
| 182 |
+
"epoch": 4.0,
|
| 183 |
+
"eval_accuracy": 0.8129496402877698,
|
| 184 |
+
"eval_loss": 0.4988514482975006,
|
| 185 |
+
"eval_runtime": 12.1672,
|
| 186 |
+
"eval_samples_per_second": 79.969,
|
| 187 |
+
"eval_steps_per_second": 5.013,
|
| 188 |
+
"step": 244
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.1,
|
| 192 |
+
"learning_rate": 8.19672131147541e-06,
|
| 193 |
+
"loss": 0.7001,
|
| 194 |
"step": 250
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 4.26,
|
| 198 |
+
"learning_rate": 8.524590163934427e-06,
|
| 199 |
+
"loss": 0.6759,
|
| 200 |
"step": 260
|
| 201 |
},
|
| 202 |
{
|
| 203 |
+
"epoch": 4.43,
|
| 204 |
+
"learning_rate": 8.852459016393443e-06,
|
| 205 |
+
"loss": 0.6793,
|
| 206 |
"step": 270
|
| 207 |
},
|
| 208 |
{
|
| 209 |
+
"epoch": 4.59,
|
| 210 |
+
"learning_rate": 9.18032786885246e-06,
|
| 211 |
+
"loss": 0.6572,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 212 |
"step": 280
|
| 213 |
},
|
| 214 |
{
|
| 215 |
+
"epoch": 4.75,
|
| 216 |
+
"learning_rate": 9.508196721311476e-06,
|
| 217 |
+
"loss": 0.6086,
|
| 218 |
"step": 290
|
| 219 |
},
|
| 220 |
{
|
| 221 |
+
"epoch": 4.92,
|
| 222 |
+
"learning_rate": 9.836065573770493e-06,
|
| 223 |
+
"loss": 0.615,
|
| 224 |
"step": 300
|
| 225 |
},
|
| 226 |
{
|
| 227 |
+
"epoch": 5.0,
|
| 228 |
+
"eval_accuracy": 0.8746145940390545,
|
| 229 |
+
"eval_loss": 0.3544515371322632,
|
| 230 |
+
"eval_runtime": 12.2633,
|
| 231 |
+
"eval_samples_per_second": 79.343,
|
| 232 |
+
"eval_steps_per_second": 4.974,
|
| 233 |
+
"step": 305
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 5.08,
|
| 237 |
+
"learning_rate": 9.836065573770493e-06,
|
| 238 |
+
"loss": 0.5495,
|
| 239 |
"step": 310
|
| 240 |
},
|
| 241 |
{
|
| 242 |
+
"epoch": 5.25,
|
| 243 |
+
"learning_rate": 9.508196721311476e-06,
|
| 244 |
+
"loss": 0.5625,
|
| 245 |
"step": 320
|
| 246 |
},
|
| 247 |
{
|
| 248 |
+
"epoch": 5.41,
|
| 249 |
+
"learning_rate": 9.18032786885246e-06,
|
| 250 |
+
"loss": 0.5274,
|
| 251 |
"step": 330
|
| 252 |
},
|
| 253 |
{
|
| 254 |
+
"epoch": 5.57,
|
| 255 |
+
"learning_rate": 8.852459016393443e-06,
|
| 256 |
+
"loss": 0.4914,
|
| 257 |
+
"step": 340
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 5.74,
|
| 261 |
+
"learning_rate": 8.524590163934427e-06,
|
| 262 |
+
"loss": 0.3984,
|
| 263 |
+
"step": 350
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 5.9,
|
| 267 |
+
"learning_rate": 8.19672131147541e-06,
|
| 268 |
+
"loss": 0.4751,
|
| 269 |
+
"step": 360
|
| 270 |
},
|
| 271 |
{
|
| 272 |
"epoch": 6.0,
|
| 273 |
+
"eval_accuracy": 0.9167523124357657,
|
| 274 |
+
"eval_loss": 0.2399262934923172,
|
| 275 |
+
"eval_runtime": 12.1531,
|
| 276 |
+
"eval_samples_per_second": 80.062,
|
| 277 |
+
"eval_steps_per_second": 5.019,
|
| 278 |
+
"step": 366
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 6.07,
|
| 282 |
+
"learning_rate": 7.868852459016394e-06,
|
| 283 |
+
"loss": 0.4427,
|
| 284 |
+
"step": 370
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 6.23,
|
| 288 |
+
"learning_rate": 7.540983606557377e-06,
|
| 289 |
+
"loss": 0.3884,
|
| 290 |
+
"step": 380
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 6.39,
|
| 294 |
+
"learning_rate": 7.213114754098361e-06,
|
| 295 |
+
"loss": 0.3789,
|
| 296 |
+
"step": 390
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 6.56,
|
| 300 |
+
"learning_rate": 6.885245901639345e-06,
|
| 301 |
+
"loss": 0.4063,
|
| 302 |
+
"step": 400
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 6.72,
|
| 306 |
+
"learning_rate": 6.5573770491803276e-06,
|
| 307 |
+
"loss": 0.3784,
|
| 308 |
+
"step": 410
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 6.89,
|
| 312 |
+
"learning_rate": 6.229508196721312e-06,
|
| 313 |
+
"loss": 0.3778,
|
| 314 |
+
"step": 420
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 7.0,
|
| 318 |
+
"eval_accuracy": 0.9558067831449126,
|
| 319 |
+
"eval_loss": 0.16275165975093842,
|
| 320 |
+
"eval_runtime": 12.3058,
|
| 321 |
+
"eval_samples_per_second": 79.068,
|
| 322 |
+
"eval_steps_per_second": 4.957,
|
| 323 |
+
"step": 427
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 7.05,
|
| 327 |
+
"learning_rate": 5.9016393442622956e-06,
|
| 328 |
+
"loss": 0.3681,
|
| 329 |
+
"step": 430
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 7.21,
|
| 333 |
+
"learning_rate": 5.573770491803278e-06,
|
| 334 |
+
"loss": 0.3397,
|
| 335 |
+
"step": 440
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 7.38,
|
| 339 |
+
"learning_rate": 5.245901639344263e-06,
|
| 340 |
+
"loss": 0.3429,
|
| 341 |
+
"step": 450
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 7.54,
|
| 345 |
+
"learning_rate": 4.918032786885246e-06,
|
| 346 |
+
"loss": 0.3384,
|
| 347 |
+
"step": 460
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 7.7,
|
| 351 |
+
"learning_rate": 4.59016393442623e-06,
|
| 352 |
+
"loss": 0.3018,
|
| 353 |
+
"step": 470
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 7.87,
|
| 357 |
+
"learning_rate": 4.2622950819672135e-06,
|
| 358 |
+
"loss": 0.3054,
|
| 359 |
+
"step": 480
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 8.0,
|
| 363 |
+
"eval_accuracy": 0.9619732785200411,
|
| 364 |
+
"eval_loss": 0.12017930299043655,
|
| 365 |
+
"eval_runtime": 12.2634,
|
| 366 |
+
"eval_samples_per_second": 79.342,
|
| 367 |
+
"eval_steps_per_second": 4.974,
|
| 368 |
+
"step": 488
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 8.03,
|
| 372 |
+
"learning_rate": 3.934426229508197e-06,
|
| 373 |
+
"loss": 0.3419,
|
| 374 |
+
"step": 490
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 8.2,
|
| 378 |
+
"learning_rate": 3.6065573770491806e-06,
|
| 379 |
+
"loss": 0.3126,
|
| 380 |
+
"step": 500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 8.36,
|
| 384 |
+
"learning_rate": 3.2786885245901638e-06,
|
| 385 |
+
"loss": 0.2932,
|
| 386 |
+
"step": 510
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 8.52,
|
| 390 |
+
"learning_rate": 2.9508196721311478e-06,
|
| 391 |
+
"loss": 0.2949,
|
| 392 |
+
"step": 520
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 8.69,
|
| 396 |
+
"learning_rate": 2.6229508196721314e-06,
|
| 397 |
+
"loss": 0.327,
|
| 398 |
+
"step": 530
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 8.85,
|
| 402 |
+
"learning_rate": 2.295081967213115e-06,
|
| 403 |
+
"loss": 0.2787,
|
| 404 |
+
"step": 540
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 9.0,
|
| 408 |
+
"eval_accuracy": 0.9732785200411099,
|
| 409 |
+
"eval_loss": 0.0988311693072319,
|
| 410 |
+
"eval_runtime": 12.0955,
|
| 411 |
+
"eval_samples_per_second": 80.443,
|
| 412 |
+
"eval_steps_per_second": 5.043,
|
| 413 |
+
"step": 549
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 9.02,
|
| 417 |
+
"learning_rate": 1.9672131147540985e-06,
|
| 418 |
+
"loss": 0.2774,
|
| 419 |
+
"step": 550
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 9.18,
|
| 423 |
+
"learning_rate": 1.6393442622950819e-06,
|
| 424 |
+
"loss": 0.2333,
|
| 425 |
+
"step": 560
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 9.34,
|
| 429 |
+
"learning_rate": 1.3114754098360657e-06,
|
| 430 |
+
"loss": 0.2573,
|
| 431 |
+
"step": 570
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 9.51,
|
| 435 |
+
"learning_rate": 9.836065573770493e-07,
|
| 436 |
+
"loss": 0.2715,
|
| 437 |
+
"step": 580
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 9.67,
|
| 441 |
+
"learning_rate": 6.557377049180328e-07,
|
| 442 |
+
"loss": 0.2168,
|
| 443 |
+
"step": 590
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 9.84,
|
| 447 |
+
"learning_rate": 3.278688524590164e-07,
|
| 448 |
+
"loss": 0.3328,
|
| 449 |
+
"step": 600
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 10.0,
|
| 453 |
+
"learning_rate": 0.0,
|
| 454 |
+
"loss": 0.253,
|
| 455 |
+
"step": 610
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 10.0,
|
| 459 |
+
"eval_accuracy": 0.9722507708119219,
|
| 460 |
+
"eval_loss": 0.09661855548620224,
|
| 461 |
+
"eval_runtime": 12.2414,
|
| 462 |
+
"eval_samples_per_second": 79.484,
|
| 463 |
+
"eval_steps_per_second": 4.983,
|
| 464 |
+
"step": 610
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 10.0,
|
| 468 |
+
"step": 610,
|
| 469 |
+
"total_flos": 8.960328656655483e+18,
|
| 470 |
+
"train_loss": 0.807003553578111,
|
| 471 |
+
"train_runtime": 6787.6894,
|
| 472 |
+
"train_samples_per_second": 5.732,
|
| 473 |
+
"train_steps_per_second": 0.09
|
| 474 |
}
|
| 475 |
],
|
| 476 |
+
"max_steps": 610,
|
| 477 |
+
"num_train_epochs": 10,
|
| 478 |
+
"total_flos": 8.960328656655483e+18,
|
| 479 |
"trial_name": null,
|
| 480 |
"trial_params": null
|
| 481 |
}
|