Commit
•
dc56829
1
Parent(s):
99c88a7
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.0013297222321853042,
|
| 5 |
+
"eval_runtime": 2.3128,
|
| 6 |
+
"eval_samples_per_second": 50.156,
|
| 7 |
+
"eval_steps_per_second": 12.539,
|
| 8 |
+
"total_flos": 5.3252144691019776e+17,
|
| 9 |
+
"train_loss": 0.2897487744918236,
|
| 10 |
+
"train_runtime": 313.9626,
|
| 11 |
+
"train_samples_per_second": 16.547,
|
| 12 |
+
"train_steps_per_second": 1.035
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy": 1.0,
|
| 4 |
+
"eval_loss": 0.0013297222321853042,
|
| 5 |
+
"eval_runtime": 2.3128,
|
| 6 |
+
"eval_samples_per_second": 50.156,
|
| 7 |
+
"eval_steps_per_second": 12.539
|
| 8 |
+
}
|
runs/Apr30_19-07-57_aa80023fdf35/events.out.tfevents.1714504578.aa80023fdf35.34.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17a5501fd1b680460338733dd2b3fd8bffc9a910e248c835e59df89c4ec4c222
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 5.3252144691019776e+17,
|
| 4 |
+
"train_loss": 0.2897487744918236,
|
| 5 |
+
"train_runtime": 313.9626,
|
| 6 |
+
"train_samples_per_second": 16.547,
|
| 7 |
+
"train_steps_per_second": 1.035
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,299 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0,
|
| 3 |
+
"best_model_checkpoint": "swinv2-base-patch4-window12to16-192to256-22kto1k-ft-finetuned-footulcer/checkpoint-325",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 325,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.15,
|
| 13 |
+
"grad_norm": 14.772844314575195,
|
| 14 |
+
"learning_rate": 1.5151515151515153e-05,
|
| 15 |
+
"loss": 0.6758,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.31,
|
| 20 |
+
"grad_norm": 17.389602661132812,
|
| 21 |
+
"learning_rate": 3.0303030303030306e-05,
|
| 22 |
+
"loss": 0.3891,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.46,
|
| 27 |
+
"grad_norm": 33.26732635498047,
|
| 28 |
+
"learning_rate": 4.545454545454546e-05,
|
| 29 |
+
"loss": 0.4548,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.62,
|
| 34 |
+
"grad_norm": 13.78071117401123,
|
| 35 |
+
"learning_rate": 4.88013698630137e-05,
|
| 36 |
+
"loss": 0.4219,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.77,
|
| 41 |
+
"grad_norm": 13.317490577697754,
|
| 42 |
+
"learning_rate": 4.708904109589041e-05,
|
| 43 |
+
"loss": 0.5415,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.92,
|
| 48 |
+
"grad_norm": 52.00563049316406,
|
| 49 |
+
"learning_rate": 4.5376712328767126e-05,
|
| 50 |
+
"loss": 0.425,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 1.0,
|
| 55 |
+
"eval_accuracy": 0.8793103448275862,
|
| 56 |
+
"eval_loss": 0.2769334614276886,
|
| 57 |
+
"eval_runtime": 2.9905,
|
| 58 |
+
"eval_samples_per_second": 38.789,
|
| 59 |
+
"eval_steps_per_second": 9.697,
|
| 60 |
+
"step": 65
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.08,
|
| 64 |
+
"grad_norm": 6.420164585113525,
|
| 65 |
+
"learning_rate": 4.366438356164384e-05,
|
| 66 |
+
"loss": 0.4124,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.23,
|
| 71 |
+
"grad_norm": 6.303807258605957,
|
| 72 |
+
"learning_rate": 4.195205479452055e-05,
|
| 73 |
+
"loss": 0.3529,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.38,
|
| 78 |
+
"grad_norm": 13.112977027893066,
|
| 79 |
+
"learning_rate": 4.0239726027397265e-05,
|
| 80 |
+
"loss": 0.3787,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.54,
|
| 85 |
+
"grad_norm": 20.180587768554688,
|
| 86 |
+
"learning_rate": 3.852739726027397e-05,
|
| 87 |
+
"loss": 0.2384,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.69,
|
| 92 |
+
"grad_norm": 19.817771911621094,
|
| 93 |
+
"learning_rate": 3.6815068493150685e-05,
|
| 94 |
+
"loss": 0.2643,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.85,
|
| 99 |
+
"grad_norm": 14.556670188903809,
|
| 100 |
+
"learning_rate": 3.51027397260274e-05,
|
| 101 |
+
"loss": 0.3526,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.0,
|
| 106 |
+
"grad_norm": 33.79677200317383,
|
| 107 |
+
"learning_rate": 3.339041095890411e-05,
|
| 108 |
+
"loss": 0.3182,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.0,
|
| 113 |
+
"eval_accuracy": 0.9827586206896551,
|
| 114 |
+
"eval_loss": 0.054682787507772446,
|
| 115 |
+
"eval_runtime": 2.2733,
|
| 116 |
+
"eval_samples_per_second": 51.027,
|
| 117 |
+
"eval_steps_per_second": 12.757,
|
| 118 |
+
"step": 130
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.15,
|
| 122 |
+
"grad_norm": 5.128925323486328,
|
| 123 |
+
"learning_rate": 3.1678082191780824e-05,
|
| 124 |
+
"loss": 0.2523,
|
| 125 |
+
"step": 140
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.31,
|
| 129 |
+
"grad_norm": 11.12570571899414,
|
| 130 |
+
"learning_rate": 2.9965753424657534e-05,
|
| 131 |
+
"loss": 0.1997,
|
| 132 |
+
"step": 150
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.46,
|
| 136 |
+
"grad_norm": 43.87520217895508,
|
| 137 |
+
"learning_rate": 2.825342465753425e-05,
|
| 138 |
+
"loss": 0.3346,
|
| 139 |
+
"step": 160
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.62,
|
| 143 |
+
"grad_norm": 7.066685199737549,
|
| 144 |
+
"learning_rate": 2.654109589041096e-05,
|
| 145 |
+
"loss": 0.3683,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.77,
|
| 150 |
+
"grad_norm": 11.333112716674805,
|
| 151 |
+
"learning_rate": 2.4828767123287673e-05,
|
| 152 |
+
"loss": 0.2765,
|
| 153 |
+
"step": 180
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.92,
|
| 157 |
+
"grad_norm": 10.573657035827637,
|
| 158 |
+
"learning_rate": 2.3116438356164386e-05,
|
| 159 |
+
"loss": 0.2053,
|
| 160 |
+
"step": 190
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 3.0,
|
| 164 |
+
"eval_accuracy": 0.9913793103448276,
|
| 165 |
+
"eval_loss": 0.028553670272231102,
|
| 166 |
+
"eval_runtime": 2.3027,
|
| 167 |
+
"eval_samples_per_second": 50.376,
|
| 168 |
+
"eval_steps_per_second": 12.594,
|
| 169 |
+
"step": 195
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 3.08,
|
| 173 |
+
"grad_norm": 9.268758773803711,
|
| 174 |
+
"learning_rate": 2.1404109589041096e-05,
|
| 175 |
+
"loss": 0.1977,
|
| 176 |
+
"step": 200
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.23,
|
| 180 |
+
"grad_norm": 62.42777633666992,
|
| 181 |
+
"learning_rate": 1.969178082191781e-05,
|
| 182 |
+
"loss": 0.3236,
|
| 183 |
+
"step": 210
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.38,
|
| 187 |
+
"grad_norm": 23.722658157348633,
|
| 188 |
+
"learning_rate": 1.7979452054794522e-05,
|
| 189 |
+
"loss": 0.1492,
|
| 190 |
+
"step": 220
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.54,
|
| 194 |
+
"grad_norm": 19.14108657836914,
|
| 195 |
+
"learning_rate": 1.6267123287671235e-05,
|
| 196 |
+
"loss": 0.2977,
|
| 197 |
+
"step": 230
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.69,
|
| 201 |
+
"grad_norm": 6.952052116394043,
|
| 202 |
+
"learning_rate": 1.4554794520547945e-05,
|
| 203 |
+
"loss": 0.2568,
|
| 204 |
+
"step": 240
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.85,
|
| 208 |
+
"grad_norm": 4.827333450317383,
|
| 209 |
+
"learning_rate": 1.284246575342466e-05,
|
| 210 |
+
"loss": 0.1743,
|
| 211 |
+
"step": 250
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 4.0,
|
| 215 |
+
"grad_norm": 3.0569567680358887,
|
| 216 |
+
"learning_rate": 1.113013698630137e-05,
|
| 217 |
+
"loss": 0.2892,
|
| 218 |
+
"step": 260
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.0,
|
| 222 |
+
"eval_accuracy": 0.9913793103448276,
|
| 223 |
+
"eval_loss": 0.01669893227517605,
|
| 224 |
+
"eval_runtime": 2.2629,
|
| 225 |
+
"eval_samples_per_second": 51.261,
|
| 226 |
+
"eval_steps_per_second": 12.815,
|
| 227 |
+
"step": 260
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 4.15,
|
| 231 |
+
"grad_norm": 16.4809627532959,
|
| 232 |
+
"learning_rate": 9.417808219178083e-06,
|
| 233 |
+
"loss": 0.1562,
|
| 234 |
+
"step": 270
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 4.31,
|
| 238 |
+
"grad_norm": 5.131158828735352,
|
| 239 |
+
"learning_rate": 7.705479452054794e-06,
|
| 240 |
+
"loss": 0.1183,
|
| 241 |
+
"step": 280
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 4.46,
|
| 245 |
+
"grad_norm": 8.427245140075684,
|
| 246 |
+
"learning_rate": 5.993150684931507e-06,
|
| 247 |
+
"loss": 0.0951,
|
| 248 |
+
"step": 290
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.62,
|
| 252 |
+
"grad_norm": 1.835396647453308,
|
| 253 |
+
"learning_rate": 4.280821917808219e-06,
|
| 254 |
+
"loss": 0.1224,
|
| 255 |
+
"step": 300
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 4.77,
|
| 259 |
+
"grad_norm": 2.6403584480285645,
|
| 260 |
+
"learning_rate": 2.5684931506849316e-06,
|
| 261 |
+
"loss": 0.1092,
|
| 262 |
+
"step": 310
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 4.92,
|
| 266 |
+
"grad_norm": 41.65308380126953,
|
| 267 |
+
"learning_rate": 8.561643835616439e-07,
|
| 268 |
+
"loss": 0.1774,
|
| 269 |
+
"step": 320
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 5.0,
|
| 273 |
+
"eval_accuracy": 1.0,
|
| 274 |
+
"eval_loss": 0.0013297222321853042,
|
| 275 |
+
"eval_runtime": 2.3045,
|
| 276 |
+
"eval_samples_per_second": 50.337,
|
| 277 |
+
"eval_steps_per_second": 12.584,
|
| 278 |
+
"step": 325
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 5.0,
|
| 282 |
+
"step": 325,
|
| 283 |
+
"total_flos": 5.3252144691019776e+17,
|
| 284 |
+
"train_loss": 0.2897487744918236,
|
| 285 |
+
"train_runtime": 313.9626,
|
| 286 |
+
"train_samples_per_second": 16.547,
|
| 287 |
+
"train_steps_per_second": 1.035
|
| 288 |
+
}
|
| 289 |
+
],
|
| 290 |
+
"logging_steps": 10,
|
| 291 |
+
"max_steps": 325,
|
| 292 |
+
"num_input_tokens_seen": 0,
|
| 293 |
+
"num_train_epochs": 5,
|
| 294 |
+
"save_steps": 500,
|
| 295 |
+
"total_flos": 5.3252144691019776e+17,
|
| 296 |
+
"train_batch_size": 4,
|
| 297 |
+
"trial_name": null,
|
| 298 |
+
"trial_params": null
|
| 299 |
+
}
|