Commit
•
91db11d
1
Parent(s):
1ea0817
End of training
Browse files- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +355 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.242021276595745,
|
| 3 |
+
"eval_accuracy": 0.7633136094674556,
|
| 4 |
+
"eval_loss": 0.875207781791687,
|
| 5 |
+
"eval_runtime": 11.2012,
|
| 6 |
+
"eval_samples_per_second": 15.088,
|
| 7 |
+
"eval_steps_per_second": 1.964
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.242021276595745,
|
| 3 |
+
"eval_accuracy": 0.7633136094674556,
|
| 4 |
+
"eval_loss": 0.875207781791687,
|
| 5 |
+
"eval_runtime": 11.2012,
|
| 6 |
+
"eval_samples_per_second": 15.088,
|
| 7 |
+
"eval_steps_per_second": 1.964
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,355 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7875,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-gesturePhasev2/checkpoint-95",
|
| 4 |
+
"epoch": 3.242021276595745,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 376,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026595744680851064,
|
| 13 |
+
"grad_norm": 13.211409568786621,
|
| 14 |
+
"learning_rate": 2.631578947368421e-06,
|
| 15 |
+
"loss": 1.594,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05319148936170213,
|
| 20 |
+
"grad_norm": 9.70760726928711,
|
| 21 |
+
"learning_rate": 5.263157894736842e-06,
|
| 22 |
+
"loss": 1.4073,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0797872340425532,
|
| 27 |
+
"grad_norm": 5.945734977722168,
|
| 28 |
+
"learning_rate": 7.894736842105265e-06,
|
| 29 |
+
"loss": 1.1347,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10638297872340426,
|
| 34 |
+
"grad_norm": 8.433605194091797,
|
| 35 |
+
"learning_rate": 9.940828402366864e-06,
|
| 36 |
+
"loss": 0.8755,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.13297872340425532,
|
| 41 |
+
"grad_norm": 5.273907661437988,
|
| 42 |
+
"learning_rate": 9.644970414201184e-06,
|
| 43 |
+
"loss": 0.6064,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1595744680851064,
|
| 48 |
+
"grad_norm": 6.961372375488281,
|
| 49 |
+
"learning_rate": 9.349112426035503e-06,
|
| 50 |
+
"loss": 0.9865,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.18617021276595744,
|
| 55 |
+
"grad_norm": 6.787417411804199,
|
| 56 |
+
"learning_rate": 9.053254437869822e-06,
|
| 57 |
+
"loss": 0.7214,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2127659574468085,
|
| 62 |
+
"grad_norm": 4.117732048034668,
|
| 63 |
+
"learning_rate": 8.757396449704143e-06,
|
| 64 |
+
"loss": 0.5468,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.2393617021276596,
|
| 69 |
+
"grad_norm": 5.041596412658691,
|
| 70 |
+
"learning_rate": 8.461538461538462e-06,
|
| 71 |
+
"loss": 0.806,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.2526595744680851,
|
| 76 |
+
"eval_accuracy": 0.7875,
|
| 77 |
+
"eval_loss": 0.8249056935310364,
|
| 78 |
+
"eval_runtime": 22.7472,
|
| 79 |
+
"eval_samples_per_second": 7.034,
|
| 80 |
+
"eval_steps_per_second": 0.879,
|
| 81 |
+
"step": 95
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.0132978723404256,
|
| 85 |
+
"grad_norm": 3.2331671714782715,
|
| 86 |
+
"learning_rate": 8.165680473372781e-06,
|
| 87 |
+
"loss": 0.6714,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.0398936170212767,
|
| 92 |
+
"grad_norm": 5.518906116485596,
|
| 93 |
+
"learning_rate": 7.869822485207102e-06,
|
| 94 |
+
"loss": 0.7709,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.0664893617021276,
|
| 99 |
+
"grad_norm": 4.540952205657959,
|
| 100 |
+
"learning_rate": 7.573964497041421e-06,
|
| 101 |
+
"loss": 0.7454,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.0930851063829787,
|
| 106 |
+
"grad_norm": 5.326303482055664,
|
| 107 |
+
"learning_rate": 7.278106508875741e-06,
|
| 108 |
+
"loss": 0.6977,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.1196808510638299,
|
| 113 |
+
"grad_norm": 9.28279972076416,
|
| 114 |
+
"learning_rate": 6.98224852071006e-06,
|
| 115 |
+
"loss": 0.6382,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.1462765957446808,
|
| 120 |
+
"grad_norm": 3.8069820404052734,
|
| 121 |
+
"learning_rate": 6.686390532544379e-06,
|
| 122 |
+
"loss": 0.7963,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.172872340425532,
|
| 127 |
+
"grad_norm": 3.917985677719116,
|
| 128 |
+
"learning_rate": 6.3905325443786995e-06,
|
| 129 |
+
"loss": 0.6782,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.199468085106383,
|
| 134 |
+
"grad_norm": 5.1802802085876465,
|
| 135 |
+
"learning_rate": 6.0946745562130185e-06,
|
| 136 |
+
"loss": 0.6922,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.226063829787234,
|
| 141 |
+
"grad_norm": 2.9730751514434814,
|
| 142 |
+
"learning_rate": 5.7988165680473375e-06,
|
| 143 |
+
"loss": 0.7682,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.252659574468085,
|
| 148 |
+
"grad_norm": 6.959865570068359,
|
| 149 |
+
"learning_rate": 5.502958579881657e-06,
|
| 150 |
+
"loss": 0.8453,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.252659574468085,
|
| 155 |
+
"eval_accuracy": 0.7875,
|
| 156 |
+
"eval_loss": 0.8071285486221313,
|
| 157 |
+
"eval_runtime": 10.9141,
|
| 158 |
+
"eval_samples_per_second": 14.66,
|
| 159 |
+
"eval_steps_per_second": 1.832,
|
| 160 |
+
"step": 190
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.026595744680851,
|
| 164 |
+
"grad_norm": 3.4485950469970703,
|
| 165 |
+
"learning_rate": 5.207100591715976e-06,
|
| 166 |
+
"loss": 0.7081,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.0531914893617023,
|
| 171 |
+
"grad_norm": 3.514798164367676,
|
| 172 |
+
"learning_rate": 4.911242603550296e-06,
|
| 173 |
+
"loss": 0.6794,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.0797872340425534,
|
| 178 |
+
"grad_norm": 4.564563751220703,
|
| 179 |
+
"learning_rate": 4.615384615384616e-06,
|
| 180 |
+
"loss": 0.8428,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.106382978723404,
|
| 185 |
+
"grad_norm": 5.189121723175049,
|
| 186 |
+
"learning_rate": 4.319526627218935e-06,
|
| 187 |
+
"loss": 0.8315,
|
| 188 |
+
"step": 230
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.132978723404255,
|
| 192 |
+
"grad_norm": 6.5788044929504395,
|
| 193 |
+
"learning_rate": 4.023668639053255e-06,
|
| 194 |
+
"loss": 0.6639,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.1595744680851063,
|
| 199 |
+
"grad_norm": 6.100628852844238,
|
| 200 |
+
"learning_rate": 3.7278106508875745e-06,
|
| 201 |
+
"loss": 0.7014,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.1861702127659575,
|
| 206 |
+
"grad_norm": 5.754176616668701,
|
| 207 |
+
"learning_rate": 3.4319526627218935e-06,
|
| 208 |
+
"loss": 0.7219,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.2127659574468086,
|
| 213 |
+
"grad_norm": 6.1187424659729,
|
| 214 |
+
"learning_rate": 3.1360946745562134e-06,
|
| 215 |
+
"loss": 0.5346,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.2393617021276597,
|
| 220 |
+
"grad_norm": 4.662169933319092,
|
| 221 |
+
"learning_rate": 2.840236686390533e-06,
|
| 222 |
+
"loss": 0.8814,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.252659574468085,
|
| 227 |
+
"eval_accuracy": 0.7875,
|
| 228 |
+
"eval_loss": 0.8307636380195618,
|
| 229 |
+
"eval_runtime": 10.9219,
|
| 230 |
+
"eval_samples_per_second": 14.649,
|
| 231 |
+
"eval_steps_per_second": 1.831,
|
| 232 |
+
"step": 285
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.0132978723404253,
|
| 236 |
+
"grad_norm": 5.648519992828369,
|
| 237 |
+
"learning_rate": 2.5443786982248527e-06,
|
| 238 |
+
"loss": 0.601,
|
| 239 |
+
"step": 290
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.0398936170212765,
|
| 243 |
+
"grad_norm": 3.8109617233276367,
|
| 244 |
+
"learning_rate": 2.2485207100591717e-06,
|
| 245 |
+
"loss": 0.618,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.0664893617021276,
|
| 250 |
+
"grad_norm": 4.005825519561768,
|
| 251 |
+
"learning_rate": 1.952662721893491e-06,
|
| 252 |
+
"loss": 0.6362,
|
| 253 |
+
"step": 310
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 3.0930851063829787,
|
| 257 |
+
"grad_norm": 6.642380237579346,
|
| 258 |
+
"learning_rate": 1.656804733727811e-06,
|
| 259 |
+
"loss": 0.8369,
|
| 260 |
+
"step": 320
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.11968085106383,
|
| 264 |
+
"grad_norm": 4.232272624969482,
|
| 265 |
+
"learning_rate": 1.3609467455621303e-06,
|
| 266 |
+
"loss": 0.7755,
|
| 267 |
+
"step": 330
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 3.146276595744681,
|
| 271 |
+
"grad_norm": 3.2232885360717773,
|
| 272 |
+
"learning_rate": 1.06508875739645e-06,
|
| 273 |
+
"loss": 0.8827,
|
| 274 |
+
"step": 340
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 3.172872340425532,
|
| 278 |
+
"grad_norm": 5.880322456359863,
|
| 279 |
+
"learning_rate": 7.692307692307694e-07,
|
| 280 |
+
"loss": 0.6081,
|
| 281 |
+
"step": 350
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.199468085106383,
|
| 285 |
+
"grad_norm": 6.9393310546875,
|
| 286 |
+
"learning_rate": 4.733727810650888e-07,
|
| 287 |
+
"loss": 0.8163,
|
| 288 |
+
"step": 360
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.226063829787234,
|
| 292 |
+
"grad_norm": 6.549450874328613,
|
| 293 |
+
"learning_rate": 1.775147928994083e-07,
|
| 294 |
+
"loss": 0.8272,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.242021276595745,
|
| 299 |
+
"eval_accuracy": 0.7875,
|
| 300 |
+
"eval_loss": 0.8259578943252563,
|
| 301 |
+
"eval_runtime": 84.7179,
|
| 302 |
+
"eval_samples_per_second": 1.889,
|
| 303 |
+
"eval_steps_per_second": 0.236,
|
| 304 |
+
"step": 376
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 3.242021276595745,
|
| 308 |
+
"step": 376,
|
| 309 |
+
"total_flos": 3.7220613152994755e+18,
|
| 310 |
+
"train_loss": 0.7830839626332546,
|
| 311 |
+
"train_runtime": 814.9801,
|
| 312 |
+
"train_samples_per_second": 3.691,
|
| 313 |
+
"train_steps_per_second": 0.461
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 3.242021276595745,
|
| 317 |
+
"eval_accuracy": 0.7633136094674556,
|
| 318 |
+
"eval_loss": 0.8752076625823975,
|
| 319 |
+
"eval_runtime": 14.0179,
|
| 320 |
+
"eval_samples_per_second": 12.056,
|
| 321 |
+
"eval_steps_per_second": 1.569,
|
| 322 |
+
"step": 376
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 3.242021276595745,
|
| 326 |
+
"eval_accuracy": 0.7633136094674556,
|
| 327 |
+
"eval_loss": 0.875207781791687,
|
| 328 |
+
"eval_runtime": 11.2012,
|
| 329 |
+
"eval_samples_per_second": 15.088,
|
| 330 |
+
"eval_steps_per_second": 1.964,
|
| 331 |
+
"step": 376
|
| 332 |
+
}
|
| 333 |
+
],
|
| 334 |
+
"logging_steps": 10,
|
| 335 |
+
"max_steps": 376,
|
| 336 |
+
"num_input_tokens_seen": 0,
|
| 337 |
+
"num_train_epochs": 9223372036854775807,
|
| 338 |
+
"save_steps": 500,
|
| 339 |
+
"stateful_callbacks": {
|
| 340 |
+
"TrainerControl": {
|
| 341 |
+
"args": {
|
| 342 |
+
"should_epoch_stop": false,
|
| 343 |
+
"should_evaluate": false,
|
| 344 |
+
"should_log": false,
|
| 345 |
+
"should_save": true,
|
| 346 |
+
"should_training_stop": true
|
| 347 |
+
},
|
| 348 |
+
"attributes": {}
|
| 349 |
+
}
|
| 350 |
+
},
|
| 351 |
+
"total_flos": 3.7220613152994755e+18,
|
| 352 |
+
"train_batch_size": 8,
|
| 353 |
+
"trial_name": null,
|
| 354 |
+
"trial_params": null
|
| 355 |
+
}
|