End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +329 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 4.878503731310727e+18,
|
| 4 |
+
"train_loss": 0.040008016563398315,
|
| 5 |
+
"train_runtime": 8568.9568,
|
| 6 |
+
"train_samples_per_second": 28.008,
|
| 7 |
+
"train_steps_per_second": 0.055
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 4.878503731310727e+18,
|
| 4 |
+
"train_loss": 0.040008016563398315,
|
| 5 |
+
"train_runtime": 8568.9568,
|
| 6 |
+
"train_samples_per_second": 28.008,
|
| 7 |
+
"train_steps_per_second": 0.055
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,329 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.024307304993271828,
|
| 3 |
+
"best_model_checkpoint": "./dinov2_LoRA_Liveness_detection_v1.1/checkpoint-448",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 64,
|
| 6 |
+
"global_step": 471,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.10191082802547771,
|
| 13 |
+
"grad_norm": 3.3717641830444336,
|
| 14 |
+
"learning_rate": 4.840764331210191e-05,
|
| 15 |
+
"loss": 0.3564,
|
| 16 |
+
"step": 16
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.20382165605095542,
|
| 20 |
+
"grad_norm": 1.4053750038146973,
|
| 21 |
+
"learning_rate": 4.6709129511677285e-05,
|
| 22 |
+
"loss": 0.1026,
|
| 23 |
+
"step": 32
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.3057324840764331,
|
| 27 |
+
"grad_norm": 1.2090115547180176,
|
| 28 |
+
"learning_rate": 4.501061571125266e-05,
|
| 29 |
+
"loss": 0.0606,
|
| 30 |
+
"step": 48
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.40764331210191085,
|
| 34 |
+
"grad_norm": 2.0192923545837402,
|
| 35 |
+
"learning_rate": 4.331210191082803e-05,
|
| 36 |
+
"loss": 0.0501,
|
| 37 |
+
"step": 64
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.40764331210191085,
|
| 41 |
+
"eval_accuracy": 0.9841349785948124,
|
| 42 |
+
"eval_f1": 0.9841073630518083,
|
| 43 |
+
"eval_loss": 0.04576801881194115,
|
| 44 |
+
"eval_precision": 0.9842311607904581,
|
| 45 |
+
"eval_recall": 0.9841349785948124,
|
| 46 |
+
"eval_runtime": 317.7291,
|
| 47 |
+
"eval_samples_per_second": 62.947,
|
| 48 |
+
"eval_steps_per_second": 7.868,
|
| 49 |
+
"step": 64
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.5095541401273885,
|
| 53 |
+
"grad_norm": 1.1693896055221558,
|
| 54 |
+
"learning_rate": 4.1613588110403404e-05,
|
| 55 |
+
"loss": 0.0437,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.6114649681528662,
|
| 60 |
+
"grad_norm": 1.2890028953552246,
|
| 61 |
+
"learning_rate": 3.991507430997877e-05,
|
| 62 |
+
"loss": 0.0472,
|
| 63 |
+
"step": 96
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.7133757961783439,
|
| 67 |
+
"grad_norm": 1.012406826019287,
|
| 68 |
+
"learning_rate": 3.821656050955414e-05,
|
| 69 |
+
"loss": 0.0388,
|
| 70 |
+
"step": 112
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.8152866242038217,
|
| 74 |
+
"grad_norm": 1.3089163303375244,
|
| 75 |
+
"learning_rate": 3.651804670912951e-05,
|
| 76 |
+
"loss": 0.0355,
|
| 77 |
+
"step": 128
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.8152866242038217,
|
| 81 |
+
"eval_accuracy": 0.9879627297909847,
|
| 82 |
+
"eval_f1": 0.9879658772630169,
|
| 83 |
+
"eval_loss": 0.034234367311000824,
|
| 84 |
+
"eval_precision": 0.9879573976357167,
|
| 85 |
+
"eval_recall": 0.9879627297909847,
|
| 86 |
+
"eval_runtime": 309.1521,
|
| 87 |
+
"eval_samples_per_second": 64.693,
|
| 88 |
+
"eval_steps_per_second": 8.087,
|
| 89 |
+
"step": 128
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.9171974522292994,
|
| 93 |
+
"grad_norm": 1.0827258825302124,
|
| 94 |
+
"learning_rate": 3.481953290870488e-05,
|
| 95 |
+
"loss": 0.0327,
|
| 96 |
+
"step": 144
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.019108280254777,
|
| 100 |
+
"grad_norm": 1.2000535726547241,
|
| 101 |
+
"learning_rate": 3.3121019108280256e-05,
|
| 102 |
+
"loss": 0.035,
|
| 103 |
+
"step": 160
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 1.1210191082802548,
|
| 107 |
+
"grad_norm": 2.487323045730591,
|
| 108 |
+
"learning_rate": 3.142250530785563e-05,
|
| 109 |
+
"loss": 0.0275,
|
| 110 |
+
"step": 176
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.2229299363057324,
|
| 114 |
+
"grad_norm": 2.1451828479766846,
|
| 115 |
+
"learning_rate": 2.9723991507431e-05,
|
| 116 |
+
"loss": 0.0253,
|
| 117 |
+
"step": 192
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.2229299363057324,
|
| 121 |
+
"eval_accuracy": 0.990078065978343,
|
| 122 |
+
"eval_f1": 0.9900620320034099,
|
| 123 |
+
"eval_loss": 0.030784847214818,
|
| 124 |
+
"eval_precision": 0.9901457841504492,
|
| 125 |
+
"eval_recall": 0.990078065978343,
|
| 126 |
+
"eval_runtime": 313.1272,
|
| 127 |
+
"eval_samples_per_second": 63.872,
|
| 128 |
+
"eval_steps_per_second": 7.984,
|
| 129 |
+
"step": 192
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.3248407643312101,
|
| 133 |
+
"grad_norm": 1.699660062789917,
|
| 134 |
+
"learning_rate": 2.802547770700637e-05,
|
| 135 |
+
"loss": 0.0241,
|
| 136 |
+
"step": 208
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.426751592356688,
|
| 140 |
+
"grad_norm": 1.5516241788864136,
|
| 141 |
+
"learning_rate": 2.632696390658174e-05,
|
| 142 |
+
"loss": 0.0186,
|
| 143 |
+
"step": 224
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.5286624203821657,
|
| 147 |
+
"grad_norm": 1.0598669052124023,
|
| 148 |
+
"learning_rate": 2.4628450106157114e-05,
|
| 149 |
+
"loss": 0.0198,
|
| 150 |
+
"step": 240
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.6305732484076434,
|
| 154 |
+
"grad_norm": 1.8472939729690552,
|
| 155 |
+
"learning_rate": 2.2929936305732484e-05,
|
| 156 |
+
"loss": 0.0226,
|
| 157 |
+
"step": 256
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.6305732484076434,
|
| 161 |
+
"eval_accuracy": 0.990984638630068,
|
| 162 |
+
"eval_f1": 0.9909855466235222,
|
| 163 |
+
"eval_loss": 0.026686813682317734,
|
| 164 |
+
"eval_precision": 0.9909829656263511,
|
| 165 |
+
"eval_recall": 0.990984638630068,
|
| 166 |
+
"eval_runtime": 311.8552,
|
| 167 |
+
"eval_samples_per_second": 64.132,
|
| 168 |
+
"eval_steps_per_second": 8.017,
|
| 169 |
+
"step": 256
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.732484076433121,
|
| 173 |
+
"grad_norm": 1.6193630695343018,
|
| 174 |
+
"learning_rate": 2.1231422505307857e-05,
|
| 175 |
+
"loss": 0.021,
|
| 176 |
+
"step": 272
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 1.8343949044585988,
|
| 180 |
+
"grad_norm": 1.1963945627212524,
|
| 181 |
+
"learning_rate": 1.953290870488323e-05,
|
| 182 |
+
"loss": 0.0196,
|
| 183 |
+
"step": 288
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 1.9363057324840764,
|
| 187 |
+
"grad_norm": 1.3636715412139893,
|
| 188 |
+
"learning_rate": 1.78343949044586e-05,
|
| 189 |
+
"loss": 0.0231,
|
| 190 |
+
"step": 304
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 2.038216560509554,
|
| 194 |
+
"grad_norm": 1.2037672996520996,
|
| 195 |
+
"learning_rate": 1.613588110403397e-05,
|
| 196 |
+
"loss": 0.0264,
|
| 197 |
+
"step": 320
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 2.038216560509554,
|
| 201 |
+
"eval_accuracy": 0.9911860992193402,
|
| 202 |
+
"eval_f1": 0.9911791189750626,
|
| 203 |
+
"eval_loss": 0.02550842985510826,
|
| 204 |
+
"eval_precision": 0.9912087608299792,
|
| 205 |
+
"eval_recall": 0.9911860992193402,
|
| 206 |
+
"eval_runtime": 310.9636,
|
| 207 |
+
"eval_samples_per_second": 64.316,
|
| 208 |
+
"eval_steps_per_second": 8.04,
|
| 209 |
+
"step": 320
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.140127388535032,
|
| 213 |
+
"grad_norm": 1.120186448097229,
|
| 214 |
+
"learning_rate": 1.4437367303609342e-05,
|
| 215 |
+
"loss": 0.0143,
|
| 216 |
+
"step": 336
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.2420382165605095,
|
| 220 |
+
"grad_norm": 1.6843891143798828,
|
| 221 |
+
"learning_rate": 1.2738853503184714e-05,
|
| 222 |
+
"loss": 0.0148,
|
| 223 |
+
"step": 352
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.343949044585987,
|
| 227 |
+
"grad_norm": 1.4458714723587036,
|
| 228 |
+
"learning_rate": 1.1040339702760085e-05,
|
| 229 |
+
"loss": 0.0179,
|
| 230 |
+
"step": 368
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.445859872611465,
|
| 234 |
+
"grad_norm": 1.239098310470581,
|
| 235 |
+
"learning_rate": 9.341825902335457e-06,
|
| 236 |
+
"loss": 0.0168,
|
| 237 |
+
"step": 384
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.445859872611465,
|
| 241 |
+
"eval_accuracy": 0.9921934021657013,
|
| 242 |
+
"eval_f1": 0.9921865805615873,
|
| 243 |
+
"eval_loss": 0.02449687570333481,
|
| 244 |
+
"eval_precision": 0.992217636683327,
|
| 245 |
+
"eval_recall": 0.9921934021657013,
|
| 246 |
+
"eval_runtime": 309.071,
|
| 247 |
+
"eval_samples_per_second": 64.71,
|
| 248 |
+
"eval_steps_per_second": 8.089,
|
| 249 |
+
"step": 384
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.5477707006369426,
|
| 253 |
+
"grad_norm": 1.4529588222503662,
|
| 254 |
+
"learning_rate": 7.643312101910828e-06,
|
| 255 |
+
"loss": 0.0152,
|
| 256 |
+
"step": 400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.6496815286624202,
|
| 260 |
+
"grad_norm": 1.1592313051223755,
|
| 261 |
+
"learning_rate": 5.9447983014862e-06,
|
| 262 |
+
"loss": 0.0127,
|
| 263 |
+
"step": 416
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.7515923566878984,
|
| 267 |
+
"grad_norm": 1.4602242708206177,
|
| 268 |
+
"learning_rate": 4.246284501061571e-06,
|
| 269 |
+
"loss": 0.0166,
|
| 270 |
+
"step": 432
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.853503184713376,
|
| 274 |
+
"grad_norm": 1.3949599266052246,
|
| 275 |
+
"learning_rate": 2.547770700636943e-06,
|
| 276 |
+
"loss": 0.0144,
|
| 277 |
+
"step": 448
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.853503184713376,
|
| 281 |
+
"eval_accuracy": 0.9917401158398388,
|
| 282 |
+
"eval_f1": 0.9917381130215306,
|
| 283 |
+
"eval_loss": 0.024307304993271828,
|
| 284 |
+
"eval_precision": 0.9917449458573591,
|
| 285 |
+
"eval_recall": 0.9917401158398388,
|
| 286 |
+
"eval_runtime": 304.7867,
|
| 287 |
+
"eval_samples_per_second": 65.62,
|
| 288 |
+
"eval_steps_per_second": 8.202,
|
| 289 |
+
"step": 448
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 2.9554140127388537,
|
| 293 |
+
"grad_norm": 0.5842750668525696,
|
| 294 |
+
"learning_rate": 8.492569002123142e-07,
|
| 295 |
+
"loss": 0.0177,
|
| 296 |
+
"step": 464
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 3.0,
|
| 300 |
+
"step": 471,
|
| 301 |
+
"total_flos": 4.878503731310727e+18,
|
| 302 |
+
"train_loss": 0.040008016563398315,
|
| 303 |
+
"train_runtime": 8568.9568,
|
| 304 |
+
"train_samples_per_second": 28.008,
|
| 305 |
+
"train_steps_per_second": 0.055
|
| 306 |
+
}
|
| 307 |
+
],
|
| 308 |
+
"logging_steps": 16,
|
| 309 |
+
"max_steps": 471,
|
| 310 |
+
"num_input_tokens_seen": 0,
|
| 311 |
+
"num_train_epochs": 3,
|
| 312 |
+
"save_steps": 64,
|
| 313 |
+
"stateful_callbacks": {
|
| 314 |
+
"TrainerControl": {
|
| 315 |
+
"args": {
|
| 316 |
+
"should_epoch_stop": false,
|
| 317 |
+
"should_evaluate": false,
|
| 318 |
+
"should_log": false,
|
| 319 |
+
"should_save": true,
|
| 320 |
+
"should_training_stop": true
|
| 321 |
+
},
|
| 322 |
+
"attributes": {}
|
| 323 |
+
}
|
| 324 |
+
},
|
| 325 |
+
"total_flos": 4.878503731310727e+18,
|
| 326 |
+
"train_batch_size": 512,
|
| 327 |
+
"trial_name": null,
|
| 328 |
+
"trial_params": null
|
| 329 |
+
}
|