🍻 cheers
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"eval_accuracy": 0.9029850746268657,
|
| 4 |
+
"eval_loss": 0.310745507478714,
|
| 5 |
+
"eval_runtime": 2.0013,
|
| 6 |
+
"eval_samples_per_second": 66.956,
|
| 7 |
+
"eval_steps_per_second": 8.494,
|
| 8 |
+
"total_flos": 4.674400602398392e+17,
|
| 9 |
+
"train_loss": 0.24295292422175407,
|
| 10 |
+
"train_runtime": 151.2132,
|
| 11 |
+
"train_samples_per_second": 39.891,
|
| 12 |
+
"train_steps_per_second": 2.539
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"eval_accuracy": 0.9029850746268657,
|
| 4 |
+
"eval_loss": 0.310745507478714,
|
| 5 |
+
"eval_runtime": 2.0013,
|
| 6 |
+
"eval_samples_per_second": 66.956,
|
| 7 |
+
"eval_steps_per_second": 8.494
|
| 8 |
+
}
|
runs/Feb08_06-08-41_8b83d831317e/events.out.tfevents.1675836810.8b83d831317e.540.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d43ec75b6654fd71bea48b6274ce7b6293af7e81d161b4be20712e96ae61a4b
|
| 3 |
+
size 363
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 8.0,
|
| 3 |
+
"total_flos": 4.674400602398392e+17,
|
| 4 |
+
"train_loss": 0.24295292422175407,
|
| 5 |
+
"train_runtime": 151.2132,
|
| 6 |
+
"train_samples_per_second": 39.891,
|
| 7 |
+
"train_steps_per_second": 2.539
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,280 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.310745507478714,
|
| 3 |
+
"best_model_checkpoint": "finetuned-vit-doc-text-classifer/checkpoint-300",
|
| 4 |
+
"epoch": 8.0,
|
| 5 |
+
"global_step": 384,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.21,
|
| 12 |
+
"learning_rate": 0.00019479166666666668,
|
| 13 |
+
"loss": 1.2187,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.42,
|
| 18 |
+
"learning_rate": 0.00018958333333333332,
|
| 19 |
+
"loss": 0.8318,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.62,
|
| 24 |
+
"learning_rate": 0.000184375,
|
| 25 |
+
"loss": 0.5961,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.83,
|
| 30 |
+
"learning_rate": 0.0001791666666666667,
|
| 31 |
+
"loss": 0.5924,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.04,
|
| 36 |
+
"learning_rate": 0.00017395833333333334,
|
| 37 |
+
"loss": 0.5361,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.25,
|
| 42 |
+
"learning_rate": 0.00016875,
|
| 43 |
+
"loss": 0.4357,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.46,
|
| 48 |
+
"learning_rate": 0.00016354166666666668,
|
| 49 |
+
"loss": 0.3754,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.67,
|
| 54 |
+
"learning_rate": 0.00015833333333333332,
|
| 55 |
+
"loss": 0.3348,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.88,
|
| 60 |
+
"learning_rate": 0.000153125,
|
| 61 |
+
"loss": 0.315,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.08,
|
| 66 |
+
"learning_rate": 0.0001484375,
|
| 67 |
+
"loss": 0.2719,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.08,
|
| 72 |
+
"eval_accuracy": 0.8656716417910447,
|
| 73 |
+
"eval_loss": 0.41197091341018677,
|
| 74 |
+
"eval_runtime": 1.3481,
|
| 75 |
+
"eval_samples_per_second": 99.4,
|
| 76 |
+
"eval_steps_per_second": 12.611,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2.29,
|
| 81 |
+
"learning_rate": 0.00014322916666666667,
|
| 82 |
+
"loss": 0.1733,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.5,
|
| 87 |
+
"learning_rate": 0.00013802083333333334,
|
| 88 |
+
"loss": 0.3338,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 2.71,
|
| 93 |
+
"learning_rate": 0.0001328125,
|
| 94 |
+
"loss": 0.2939,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.92,
|
| 99 |
+
"learning_rate": 0.00012760416666666668,
|
| 100 |
+
"loss": 0.3062,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 3.12,
|
| 105 |
+
"learning_rate": 0.00012239583333333332,
|
| 106 |
+
"loss": 0.203,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 3.33,
|
| 111 |
+
"learning_rate": 0.00011718750000000001,
|
| 112 |
+
"loss": 0.1359,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.54,
|
| 117 |
+
"learning_rate": 0.00011197916666666668,
|
| 118 |
+
"loss": 0.1954,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 3.75,
|
| 123 |
+
"learning_rate": 0.00010677083333333334,
|
| 124 |
+
"loss": 0.123,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.96,
|
| 129 |
+
"learning_rate": 0.00010156250000000001,
|
| 130 |
+
"loss": 0.1509,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 4.17,
|
| 135 |
+
"learning_rate": 9.635416666666667e-05,
|
| 136 |
+
"loss": 0.1027,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 4.17,
|
| 141 |
+
"eval_accuracy": 0.8880597014925373,
|
| 142 |
+
"eval_loss": 0.3907298743724823,
|
| 143 |
+
"eval_runtime": 2.1054,
|
| 144 |
+
"eval_samples_per_second": 63.647,
|
| 145 |
+
"eval_steps_per_second": 8.075,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 4.38,
|
| 150 |
+
"learning_rate": 9.114583333333334e-05,
|
| 151 |
+
"loss": 0.1048,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.58,
|
| 156 |
+
"learning_rate": 8.593750000000001e-05,
|
| 157 |
+
"loss": 0.1449,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 4.79,
|
| 162 |
+
"learning_rate": 8.072916666666667e-05,
|
| 163 |
+
"loss": 0.1575,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 5.0,
|
| 168 |
+
"learning_rate": 7.552083333333334e-05,
|
| 169 |
+
"loss": 0.267,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 5.21,
|
| 174 |
+
"learning_rate": 7.031250000000001e-05,
|
| 175 |
+
"loss": 0.1269,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 5.42,
|
| 180 |
+
"learning_rate": 6.510416666666667e-05,
|
| 181 |
+
"loss": 0.0954,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 5.62,
|
| 186 |
+
"learning_rate": 5.989583333333334e-05,
|
| 187 |
+
"loss": 0.0505,
|
| 188 |
+
"step": 270
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 5.83,
|
| 192 |
+
"learning_rate": 5.46875e-05,
|
| 193 |
+
"loss": 0.0379,
|
| 194 |
+
"step": 280
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 6.04,
|
| 198 |
+
"learning_rate": 4.947916666666667e-05,
|
| 199 |
+
"loss": 0.13,
|
| 200 |
+
"step": 290
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 6.25,
|
| 204 |
+
"learning_rate": 4.4270833333333337e-05,
|
| 205 |
+
"loss": 0.0723,
|
| 206 |
+
"step": 300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 6.25,
|
| 210 |
+
"eval_accuracy": 0.9029850746268657,
|
| 211 |
+
"eval_loss": 0.310745507478714,
|
| 212 |
+
"eval_runtime": 1.3337,
|
| 213 |
+
"eval_samples_per_second": 100.474,
|
| 214 |
+
"eval_steps_per_second": 12.747,
|
| 215 |
+
"step": 300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 6.46,
|
| 219 |
+
"learning_rate": 3.90625e-05,
|
| 220 |
+
"loss": 0.1076,
|
| 221 |
+
"step": 310
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 6.67,
|
| 225 |
+
"learning_rate": 3.385416666666667e-05,
|
| 226 |
+
"loss": 0.1555,
|
| 227 |
+
"step": 320
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 6.88,
|
| 231 |
+
"learning_rate": 2.8645833333333333e-05,
|
| 232 |
+
"loss": 0.0671,
|
| 233 |
+
"step": 330
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 7.08,
|
| 237 |
+
"learning_rate": 2.34375e-05,
|
| 238 |
+
"loss": 0.082,
|
| 239 |
+
"step": 340
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 7.29,
|
| 243 |
+
"learning_rate": 1.8229166666666668e-05,
|
| 244 |
+
"loss": 0.0557,
|
| 245 |
+
"step": 350
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 7.5,
|
| 249 |
+
"learning_rate": 1.3020833333333334e-05,
|
| 250 |
+
"loss": 0.0301,
|
| 251 |
+
"step": 360
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 7.71,
|
| 255 |
+
"learning_rate": 7.8125e-06,
|
| 256 |
+
"loss": 0.0318,
|
| 257 |
+
"step": 370
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 7.92,
|
| 261 |
+
"learning_rate": 2.604166666666667e-06,
|
| 262 |
+
"loss": 0.0667,
|
| 263 |
+
"step": 380
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 8.0,
|
| 267 |
+
"step": 384,
|
| 268 |
+
"total_flos": 4.674400602398392e+17,
|
| 269 |
+
"train_loss": 0.24295292422175407,
|
| 270 |
+
"train_runtime": 151.2132,
|
| 271 |
+
"train_samples_per_second": 39.891,
|
| 272 |
+
"train_steps_per_second": 2.539
|
| 273 |
+
}
|
| 274 |
+
],
|
| 275 |
+
"max_steps": 384,
|
| 276 |
+
"num_train_epochs": 8,
|
| 277 |
+
"total_flos": 4.674400602398392e+17,
|
| 278 |
+
"trial_name": null,
|
| 279 |
+
"trial_params": null
|
| 280 |
+
}
|