Commit
•
1a2c9af
1
Parent(s):
7b08026
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 80.0,
|
| 3 |
+
"eval_accuracy": 0.9559725730783111,
|
| 4 |
+
"eval_loss": 0.19736222922801971,
|
| 5 |
+
"eval_runtime": 64.1732,
|
| 6 |
+
"eval_samples_per_second": 86.36,
|
| 7 |
+
"eval_steps_per_second": 2.711,
|
| 8 |
+
"total_flos": 8.01589959270506e+19,
|
| 9 |
+
"train_loss": 0.12853888580114534,
|
| 10 |
+
"train_runtime": 24615.3911,
|
| 11 |
+
"train_samples_per_second": 42.022,
|
| 12 |
+
"train_steps_per_second": 0.328
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 80.0,
|
| 3 |
+
"eval_accuracy": 0.9559725730783111,
|
| 4 |
+
"eval_loss": 0.19736222922801971,
|
| 5 |
+
"eval_runtime": 64.1732,
|
| 6 |
+
"eval_samples_per_second": 86.36,
|
| 7 |
+
"eval_steps_per_second": 2.711
|
| 8 |
+
}
|
runs/Apr03_16-48-58_271b37cd814f/events.out.tfevents.1649029544.271b37cd814f.97.4
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:663ec072a9db5fa6d487389fe88bf81fe2b4f4b1ef189b585595e55bcb15dda8
|
| 3 |
+
size 363
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 80.0,
|
| 3 |
+
"total_flos": 8.01589959270506e+19,
|
| 4 |
+
"train_loss": 0.12853888580114534,
|
| 5 |
+
"train_runtime": 24615.3911,
|
| 6 |
+
"train_samples_per_second": 42.022,
|
| 7 |
+
"train_steps_per_second": 0.328
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,265 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9559725730783111,
|
| 3 |
+
"best_model_checkpoint": "violation-classification-bantai-vit-v80ep/checkpoint-8000",
|
| 4 |
+
"epoch": 79.99753086419753,
|
| 5 |
+
"global_step": 8080,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 4.95,
|
| 12 |
+
"learning_rate": 3.094059405940594e-05,
|
| 13 |
+
"loss": 0.797,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 4.95,
|
| 18 |
+
"eval_accuracy": 0.8715265247203176,
|
| 19 |
+
"eval_loss": 0.392585426568985,
|
| 20 |
+
"eval_runtime": 63.7974,
|
| 21 |
+
"eval_samples_per_second": 86.869,
|
| 22 |
+
"eval_steps_per_second": 2.727,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 9.9,
|
| 27 |
+
"learning_rate": 4.867986798679868e-05,
|
| 28 |
+
"loss": 0.3095,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 9.9,
|
| 33 |
+
"eval_accuracy": 0.9106820642367376,
|
| 34 |
+
"eval_loss": 0.2596624493598938,
|
| 35 |
+
"eval_runtime": 63.8682,
|
| 36 |
+
"eval_samples_per_second": 86.772,
|
| 37 |
+
"eval_steps_per_second": 2.724,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 14.85,
|
| 42 |
+
"learning_rate": 4.5242024202420245e-05,
|
| 43 |
+
"loss": 0.1726,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 14.85,
|
| 48 |
+
"eval_accuracy": 0.9252977264525442,
|
| 49 |
+
"eval_loss": 0.2156512439250946,
|
| 50 |
+
"eval_runtime": 63.7151,
|
| 51 |
+
"eval_samples_per_second": 86.981,
|
| 52 |
+
"eval_steps_per_second": 2.731,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 19.8,
|
| 57 |
+
"learning_rate": 4.18041804180418e-05,
|
| 58 |
+
"loss": 0.1259,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 19.8,
|
| 63 |
+
"eval_accuracy": 0.939191627571274,
|
| 64 |
+
"eval_loss": 0.1869734823703766,
|
| 65 |
+
"eval_runtime": 63.965,
|
| 66 |
+
"eval_samples_per_second": 86.641,
|
| 67 |
+
"eval_steps_per_second": 2.72,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 24.75,
|
| 72 |
+
"learning_rate": 3.8366336633663367e-05,
|
| 73 |
+
"loss": 0.0959,
|
| 74 |
+
"step": 2500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 24.75,
|
| 78 |
+
"eval_accuracy": 0.9444243955250812,
|
| 79 |
+
"eval_loss": 0.17974421381950378,
|
| 80 |
+
"eval_runtime": 63.8189,
|
| 81 |
+
"eval_samples_per_second": 86.84,
|
| 82 |
+
"eval_steps_per_second": 2.726,
|
| 83 |
+
"step": 2500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 29.7,
|
| 87 |
+
"learning_rate": 3.492849284928493e-05,
|
| 88 |
+
"loss": 0.0835,
|
| 89 |
+
"step": 3000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 29.7,
|
| 93 |
+
"eval_accuracy": 0.9354023818116204,
|
| 94 |
+
"eval_loss": 0.22931958734989166,
|
| 95 |
+
"eval_runtime": 63.8202,
|
| 96 |
+
"eval_samples_per_second": 86.838,
|
| 97 |
+
"eval_steps_per_second": 2.726,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 34.65,
|
| 102 |
+
"learning_rate": 3.149064906490649e-05,
|
| 103 |
+
"loss": 0.0722,
|
| 104 |
+
"step": 3500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 34.65,
|
| 108 |
+
"eval_accuracy": 0.9440635149765427,
|
| 109 |
+
"eval_loss": 0.19207924604415894,
|
| 110 |
+
"eval_runtime": 63.813,
|
| 111 |
+
"eval_samples_per_second": 86.848,
|
| 112 |
+
"eval_steps_per_second": 2.727,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 39.6,
|
| 117 |
+
"learning_rate": 2.8052805280528056e-05,
|
| 118 |
+
"loss": 0.0628,
|
| 119 |
+
"step": 4000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 39.6,
|
| 123 |
+
"eval_accuracy": 0.9491158426560808,
|
| 124 |
+
"eval_loss": 0.18968293070793152,
|
| 125 |
+
"eval_runtime": 63.6858,
|
| 126 |
+
"eval_samples_per_second": 87.021,
|
| 127 |
+
"eval_steps_per_second": 2.732,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 44.55,
|
| 132 |
+
"learning_rate": 2.4614961496149617e-05,
|
| 133 |
+
"loss": 0.059,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 44.55,
|
| 138 |
+
"eval_accuracy": 0.9520028870443883,
|
| 139 |
+
"eval_loss": 0.1719067543745041,
|
| 140 |
+
"eval_runtime": 63.8275,
|
| 141 |
+
"eval_samples_per_second": 86.828,
|
| 142 |
+
"eval_steps_per_second": 2.726,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 49.5,
|
| 147 |
+
"learning_rate": 2.1177117711771178e-05,
|
| 148 |
+
"loss": 0.0531,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 49.5,
|
| 153 |
+
"eval_accuracy": 0.9512811259473114,
|
| 154 |
+
"eval_loss": 0.19867293536663055,
|
| 155 |
+
"eval_runtime": 63.7873,
|
| 156 |
+
"eval_samples_per_second": 86.882,
|
| 157 |
+
"eval_steps_per_second": 2.728,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 54.45,
|
| 162 |
+
"learning_rate": 1.7739273927392743e-05,
|
| 163 |
+
"loss": 0.046,
|
| 164 |
+
"step": 5500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 54.45,
|
| 168 |
+
"eval_accuracy": 0.9556116925297726,
|
| 169 |
+
"eval_loss": 0.17134784162044525,
|
| 170 |
+
"eval_runtime": 63.8123,
|
| 171 |
+
"eval_samples_per_second": 86.848,
|
| 172 |
+
"eval_steps_per_second": 2.727,
|
| 173 |
+
"step": 5500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 59.4,
|
| 177 |
+
"learning_rate": 1.4301430143014302e-05,
|
| 178 |
+
"loss": 0.0444,
|
| 179 |
+
"step": 6000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 59.4,
|
| 183 |
+
"eval_accuracy": 0.952544207867196,
|
| 184 |
+
"eval_loss": 0.2016247808933258,
|
| 185 |
+
"eval_runtime": 63.9341,
|
| 186 |
+
"eval_samples_per_second": 86.683,
|
| 187 |
+
"eval_steps_per_second": 2.722,
|
| 188 |
+
"step": 6000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 64.36,
|
| 192 |
+
"learning_rate": 1.0863586358635865e-05,
|
| 193 |
+
"loss": 0.042,
|
| 194 |
+
"step": 6500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 64.36,
|
| 198 |
+
"eval_accuracy": 0.952544207867196,
|
| 199 |
+
"eval_loss": 0.19500622153282166,
|
| 200 |
+
"eval_runtime": 63.6967,
|
| 201 |
+
"eval_samples_per_second": 87.006,
|
| 202 |
+
"eval_steps_per_second": 2.732,
|
| 203 |
+
"step": 6500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 69.31,
|
| 207 |
+
"learning_rate": 7.4257425742574256e-06,
|
| 208 |
+
"loss": 0.0363,
|
| 209 |
+
"step": 7000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 69.31,
|
| 213 |
+
"eval_accuracy": 0.9548899314326957,
|
| 214 |
+
"eval_loss": 0.20167508721351624,
|
| 215 |
+
"eval_runtime": 63.6587,
|
| 216 |
+
"eval_samples_per_second": 87.058,
|
| 217 |
+
"eval_steps_per_second": 2.733,
|
| 218 |
+
"step": 7000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 74.26,
|
| 222 |
+
"learning_rate": 3.987898789878988e-06,
|
| 223 |
+
"loss": 0.037,
|
| 224 |
+
"step": 7500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 74.26,
|
| 228 |
+
"eval_accuracy": 0.955070371706965,
|
| 229 |
+
"eval_loss": 0.1943303495645523,
|
| 230 |
+
"eval_runtime": 63.8065,
|
| 231 |
+
"eval_samples_per_second": 86.856,
|
| 232 |
+
"eval_steps_per_second": 2.727,
|
| 233 |
+
"step": 7500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 79.21,
|
| 237 |
+
"learning_rate": 5.500550055005501e-07,
|
| 238 |
+
"loss": 0.0343,
|
| 239 |
+
"step": 8000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 79.21,
|
| 243 |
+
"eval_accuracy": 0.9559725730783111,
|
| 244 |
+
"eval_loss": 0.19736222922801971,
|
| 245 |
+
"eval_runtime": 63.8534,
|
| 246 |
+
"eval_samples_per_second": 86.793,
|
| 247 |
+
"eval_steps_per_second": 2.725,
|
| 248 |
+
"step": 8000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 80.0,
|
| 252 |
+
"step": 8080,
|
| 253 |
+
"total_flos": 8.01589959270506e+19,
|
| 254 |
+
"train_loss": 0.12853888580114534,
|
| 255 |
+
"train_runtime": 24615.3911,
|
| 256 |
+
"train_samples_per_second": 42.022,
|
| 257 |
+
"train_steps_per_second": 0.328
|
| 258 |
+
}
|
| 259 |
+
],
|
| 260 |
+
"max_steps": 8080,
|
| 261 |
+
"num_train_epochs": 80,
|
| 262 |
+
"total_flos": 8.01589959270506e+19,
|
| 263 |
+
"trial_name": null,
|
| 264 |
+
"trial_params": null
|
| 265 |
+
}
|