End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +327 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 3.7851898112504986e+18,
|
| 4 |
+
"train_loss": 0.016959682344506035,
|
| 5 |
+
"train_runtime": 15597.9468,
|
| 6 |
+
"train_samples_per_second": 48.724,
|
| 7 |
+
"train_steps_per_second": 0.063
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 3.7851898112504986e+18,
|
| 4 |
+
"train_loss": 0.016959682344506035,
|
| 5 |
+
"train_runtime": 15597.9468,
|
| 6 |
+
"train_samples_per_second": 48.724,
|
| 7 |
+
"train_steps_per_second": 0.063
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,327 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.012290850281715393,
|
| 3 |
+
"best_model_checkpoint": "./dinov2_Liveness_detection_v2.1.3/checkpoint-960",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 64,
|
| 6 |
+
"global_step": 990,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.32323232323232326,
|
| 13 |
+
"grad_norm": 4.5385613441467285,
|
| 14 |
+
"learning_rate": 4.6919191919191926e-05,
|
| 15 |
+
"loss": 0.1276,
|
| 16 |
+
"step": 64
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.32323232323232326,
|
| 20 |
+
"eval_accuracy": 0.992,
|
| 21 |
+
"eval_f1": 0.9919817787996676,
|
| 22 |
+
"eval_loss": 0.02389722876250744,
|
| 23 |
+
"eval_precision": 0.9921102809473272,
|
| 24 |
+
"eval_recall": 0.992,
|
| 25 |
+
"eval_runtime": 460.3117,
|
| 26 |
+
"eval_samples_per_second": 82.553,
|
| 27 |
+
"eval_steps_per_second": 10.319,
|
| 28 |
+
"step": 64
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.6464646464646465,
|
| 32 |
+
"grad_norm": 4.910226345062256,
|
| 33 |
+
"learning_rate": 4.368686868686869e-05,
|
| 34 |
+
"loss": 0.0273,
|
| 35 |
+
"step": 128
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.6464646464646465,
|
| 39 |
+
"eval_accuracy": 0.9907894736842106,
|
| 40 |
+
"eval_f1": 0.9908113833952961,
|
| 41 |
+
"eval_loss": 0.02526354417204857,
|
| 42 |
+
"eval_precision": 0.9908335389427423,
|
| 43 |
+
"eval_recall": 0.9907894736842106,
|
| 44 |
+
"eval_runtime": 452.9825,
|
| 45 |
+
"eval_samples_per_second": 83.888,
|
| 46 |
+
"eval_steps_per_second": 10.486,
|
| 47 |
+
"step": 128
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.9696969696969697,
|
| 51 |
+
"grad_norm": 5.558176040649414,
|
| 52 |
+
"learning_rate": 4.045454545454546e-05,
|
| 53 |
+
"loss": 0.0236,
|
| 54 |
+
"step": 192
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 0.9696969696969697,
|
| 58 |
+
"eval_accuracy": 0.9908157894736842,
|
| 59 |
+
"eval_f1": 0.9908335841677582,
|
| 60 |
+
"eval_loss": 0.025667043402791023,
|
| 61 |
+
"eval_precision": 0.9908379939854149,
|
| 62 |
+
"eval_recall": 0.9908157894736842,
|
| 63 |
+
"eval_runtime": 436.548,
|
| 64 |
+
"eval_samples_per_second": 87.047,
|
| 65 |
+
"eval_steps_per_second": 10.881,
|
| 66 |
+
"step": 192
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.2929292929292928,
|
| 70 |
+
"grad_norm": 4.217276096343994,
|
| 71 |
+
"learning_rate": 3.722222222222222e-05,
|
| 72 |
+
"loss": 0.015,
|
| 73 |
+
"step": 256
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.2929292929292928,
|
| 77 |
+
"eval_accuracy": 0.9936052631578948,
|
| 78 |
+
"eval_f1": 0.9935969863198886,
|
| 79 |
+
"eval_loss": 0.022336727008223534,
|
| 80 |
+
"eval_precision": 0.993646144284306,
|
| 81 |
+
"eval_recall": 0.9936052631578948,
|
| 82 |
+
"eval_runtime": 411.4161,
|
| 83 |
+
"eval_samples_per_second": 92.364,
|
| 84 |
+
"eval_steps_per_second": 11.545,
|
| 85 |
+
"step": 256
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 1.6161616161616161,
|
| 89 |
+
"grad_norm": 5.249541282653809,
|
| 90 |
+
"learning_rate": 3.398989898989899e-05,
|
| 91 |
+
"loss": 0.0133,
|
| 92 |
+
"step": 320
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.6161616161616161,
|
| 96 |
+
"eval_accuracy": 0.9953684210526316,
|
| 97 |
+
"eval_f1": 0.9953655616272311,
|
| 98 |
+
"eval_loss": 0.014436695724725723,
|
| 99 |
+
"eval_precision": 0.9953797576759101,
|
| 100 |
+
"eval_recall": 0.9953684210526316,
|
| 101 |
+
"eval_runtime": 427.4015,
|
| 102 |
+
"eval_samples_per_second": 88.909,
|
| 103 |
+
"eval_steps_per_second": 11.114,
|
| 104 |
+
"step": 320
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.9393939393939394,
|
| 108 |
+
"grad_norm": 2.3672614097595215,
|
| 109 |
+
"learning_rate": 3.0757575757575755e-05,
|
| 110 |
+
"loss": 0.0149,
|
| 111 |
+
"step": 384
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 1.9393939393939394,
|
| 115 |
+
"eval_accuracy": 0.9912894736842105,
|
| 116 |
+
"eval_f1": 0.991312990658771,
|
| 117 |
+
"eval_loss": 0.027096593752503395,
|
| 118 |
+
"eval_precision": 0.9913564591011076,
|
| 119 |
+
"eval_recall": 0.9912894736842105,
|
| 120 |
+
"eval_runtime": 433.6926,
|
| 121 |
+
"eval_samples_per_second": 87.62,
|
| 122 |
+
"eval_steps_per_second": 10.952,
|
| 123 |
+
"step": 384
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 2.2626262626262625,
|
| 127 |
+
"grad_norm": 5.466893672943115,
|
| 128 |
+
"learning_rate": 2.7525252525252528e-05,
|
| 129 |
+
"loss": 0.0097,
|
| 130 |
+
"step": 448
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 2.2626262626262625,
|
| 134 |
+
"eval_accuracy": 0.9921842105263158,
|
| 135 |
+
"eval_f1": 0.9921762186763253,
|
| 136 |
+
"eval_loss": 0.023379703983664513,
|
| 137 |
+
"eval_precision": 0.992215426282248,
|
| 138 |
+
"eval_recall": 0.9921842105263158,
|
| 139 |
+
"eval_runtime": 433.2705,
|
| 140 |
+
"eval_samples_per_second": 87.705,
|
| 141 |
+
"eval_steps_per_second": 10.963,
|
| 142 |
+
"step": 448
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.5858585858585856,
|
| 146 |
+
"grad_norm": 3.6617612838745117,
|
| 147 |
+
"learning_rate": 2.4292929292929294e-05,
|
| 148 |
+
"loss": 0.009,
|
| 149 |
+
"step": 512
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 2.5858585858585856,
|
| 153 |
+
"eval_accuracy": 0.9953947368421052,
|
| 154 |
+
"eval_f1": 0.9953943124709226,
|
| 155 |
+
"eval_loss": 0.014852975495159626,
|
| 156 |
+
"eval_precision": 0.99539571198968,
|
| 157 |
+
"eval_recall": 0.9953947368421052,
|
| 158 |
+
"eval_runtime": 424.4172,
|
| 159 |
+
"eval_samples_per_second": 89.535,
|
| 160 |
+
"eval_steps_per_second": 11.192,
|
| 161 |
+
"step": 512
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 2.909090909090909,
|
| 165 |
+
"grad_norm": 6.237107753753662,
|
| 166 |
+
"learning_rate": 2.106060606060606e-05,
|
| 167 |
+
"loss": 0.0076,
|
| 168 |
+
"step": 576
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 2.909090909090909,
|
| 172 |
+
"eval_accuracy": 0.9951578947368421,
|
| 173 |
+
"eval_f1": 0.9951515767526712,
|
| 174 |
+
"eval_loss": 0.018406108021736145,
|
| 175 |
+
"eval_precision": 0.995195298134365,
|
| 176 |
+
"eval_recall": 0.9951578947368421,
|
| 177 |
+
"eval_runtime": 429.5702,
|
| 178 |
+
"eval_samples_per_second": 88.461,
|
| 179 |
+
"eval_steps_per_second": 11.058,
|
| 180 |
+
"step": 576
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 3.2323232323232323,
|
| 184 |
+
"grad_norm": 2.1357200145721436,
|
| 185 |
+
"learning_rate": 1.782828282828283e-05,
|
| 186 |
+
"loss": 0.0045,
|
| 187 |
+
"step": 640
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 3.2323232323232323,
|
| 191 |
+
"eval_accuracy": 0.9950526315789474,
|
| 192 |
+
"eval_f1": 0.995057445012908,
|
| 193 |
+
"eval_loss": 0.02012482099235058,
|
| 194 |
+
"eval_precision": 0.9950573808148804,
|
| 195 |
+
"eval_recall": 0.9950526315789474,
|
| 196 |
+
"eval_runtime": 425.5529,
|
| 197 |
+
"eval_samples_per_second": 89.296,
|
| 198 |
+
"eval_steps_per_second": 11.162,
|
| 199 |
+
"step": 640
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 3.5555555555555554,
|
| 203 |
+
"grad_norm": 2.942300319671631,
|
| 204 |
+
"learning_rate": 1.4595959595959597e-05,
|
| 205 |
+
"loss": 0.0032,
|
| 206 |
+
"step": 704
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 3.5555555555555554,
|
| 210 |
+
"eval_accuracy": 0.9958421052631579,
|
| 211 |
+
"eval_f1": 0.9958409965664561,
|
| 212 |
+
"eval_loss": 0.01694703847169876,
|
| 213 |
+
"eval_precision": 0.9958453781926115,
|
| 214 |
+
"eval_recall": 0.9958421052631579,
|
| 215 |
+
"eval_runtime": 425.2456,
|
| 216 |
+
"eval_samples_per_second": 89.36,
|
| 217 |
+
"eval_steps_per_second": 11.17,
|
| 218 |
+
"step": 704
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.878787878787879,
|
| 222 |
+
"grad_norm": 3.011981964111328,
|
| 223 |
+
"learning_rate": 1.1363636363636365e-05,
|
| 224 |
+
"loss": 0.0029,
|
| 225 |
+
"step": 768
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.878787878787879,
|
| 229 |
+
"eval_accuracy": 0.9960526315789474,
|
| 230 |
+
"eval_f1": 0.9960485541258135,
|
| 231 |
+
"eval_loss": 0.017786145210266113,
|
| 232 |
+
"eval_precision": 0.9960764049461148,
|
| 233 |
+
"eval_recall": 0.9960526315789474,
|
| 234 |
+
"eval_runtime": 424.6937,
|
| 235 |
+
"eval_samples_per_second": 89.476,
|
| 236 |
+
"eval_steps_per_second": 11.185,
|
| 237 |
+
"step": 768
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 4.202020202020202,
|
| 241 |
+
"grad_norm": 1.3001554012298584,
|
| 242 |
+
"learning_rate": 8.131313131313131e-06,
|
| 243 |
+
"loss": 0.002,
|
| 244 |
+
"step": 832
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 4.202020202020202,
|
| 248 |
+
"eval_accuracy": 0.9968684210526316,
|
| 249 |
+
"eval_f1": 0.9968690924275427,
|
| 250 |
+
"eval_loss": 0.01477312296628952,
|
| 251 |
+
"eval_precision": 0.9968677678038259,
|
| 252 |
+
"eval_recall": 0.9968684210526316,
|
| 253 |
+
"eval_runtime": 434.9942,
|
| 254 |
+
"eval_samples_per_second": 87.357,
|
| 255 |
+
"eval_steps_per_second": 10.92,
|
| 256 |
+
"step": 832
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 4.525252525252525,
|
| 260 |
+
"grad_norm": 0.022571153938770294,
|
| 261 |
+
"learning_rate": 4.8989898989899e-06,
|
| 262 |
+
"loss": 0.001,
|
| 263 |
+
"step": 896
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 4.525252525252525,
|
| 267 |
+
"eval_accuracy": 0.9973421052631579,
|
| 268 |
+
"eval_f1": 0.9973425125167297,
|
| 269 |
+
"eval_loss": 0.013463860377669334,
|
| 270 |
+
"eval_precision": 0.99734164275383,
|
| 271 |
+
"eval_recall": 0.9973421052631579,
|
| 272 |
+
"eval_runtime": 426.0296,
|
| 273 |
+
"eval_samples_per_second": 89.196,
|
| 274 |
+
"eval_steps_per_second": 11.149,
|
| 275 |
+
"step": 896
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 4.848484848484849,
|
| 279 |
+
"grad_norm": 0.5426102876663208,
|
| 280 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 281 |
+
"loss": 0.0007,
|
| 282 |
+
"step": 960
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 4.848484848484849,
|
| 286 |
+
"eval_accuracy": 0.997578947368421,
|
| 287 |
+
"eval_f1": 0.9975778029867857,
|
| 288 |
+
"eval_loss": 0.012290850281715393,
|
| 289 |
+
"eval_precision": 0.9975845474538715,
|
| 290 |
+
"eval_recall": 0.997578947368421,
|
| 291 |
+
"eval_runtime": 437.0403,
|
| 292 |
+
"eval_samples_per_second": 86.949,
|
| 293 |
+
"eval_steps_per_second": 10.869,
|
| 294 |
+
"step": 960
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 5.0,
|
| 298 |
+
"step": 990,
|
| 299 |
+
"total_flos": 3.7851898112504986e+18,
|
| 300 |
+
"train_loss": 0.016959682344506035,
|
| 301 |
+
"train_runtime": 15597.9468,
|
| 302 |
+
"train_samples_per_second": 48.724,
|
| 303 |
+
"train_steps_per_second": 0.063
|
| 304 |
+
}
|
| 305 |
+
],
|
| 306 |
+
"logging_steps": 64,
|
| 307 |
+
"max_steps": 990,
|
| 308 |
+
"num_input_tokens_seen": 0,
|
| 309 |
+
"num_train_epochs": 5,
|
| 310 |
+
"save_steps": 64,
|
| 311 |
+
"stateful_callbacks": {
|
| 312 |
+
"TrainerControl": {
|
| 313 |
+
"args": {
|
| 314 |
+
"should_epoch_stop": false,
|
| 315 |
+
"should_evaluate": false,
|
| 316 |
+
"should_log": false,
|
| 317 |
+
"should_save": true,
|
| 318 |
+
"should_training_stop": true
|
| 319 |
+
},
|
| 320 |
+
"attributes": {}
|
| 321 |
+
}
|
| 322 |
+
},
|
| 323 |
+
"total_flos": 3.7851898112504986e+18,
|
| 324 |
+
"train_batch_size": 768,
|
| 325 |
+
"trial_name": null,
|
| 326 |
+
"trial_params": null
|
| 327 |
+
}
|