End of training
Browse files- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +306 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.06,
|
| 3 |
+
"eval_accuracy": 0.5011600928074246,
|
| 4 |
+
"eval_loss": 0.8047754168510437,
|
| 5 |
+
"eval_runtime": 160.5206,
|
| 6 |
+
"eval_samples_per_second": 5.37,
|
| 7 |
+
"eval_steps_per_second": 0.336
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.06,
|
| 3 |
+
"eval_accuracy": 0.5011600928074246,
|
| 4 |
+
"eval_loss": 0.8047754168510437,
|
| 5 |
+
"eval_runtime": 160.5206,
|
| 6 |
+
"eval_samples_per_second": 5.37,
|
| 7 |
+
"eval_steps_per_second": 0.336
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,306 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5313496280552603,
|
| 3 |
+
"best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-ElderReact-anger-balanced/checkpoint-75",
|
| 4 |
+
"epoch": 9.0625,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 240,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.04,
|
| 13 |
+
"grad_norm": 10.033487319946289,
|
| 14 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 15 |
+
"loss": 0.7267,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.08,
|
| 20 |
+
"grad_norm": 10.82132339477539,
|
| 21 |
+
"learning_rate": 4.166666666666667e-05,
|
| 22 |
+
"loss": 0.6923,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.1,
|
| 27 |
+
"eval_accuracy": 0.46014877789585545,
|
| 28 |
+
"eval_loss": 0.8338257670402527,
|
| 29 |
+
"eval_runtime": 193.5219,
|
| 30 |
+
"eval_samples_per_second": 4.862,
|
| 31 |
+
"eval_steps_per_second": 0.305,
|
| 32 |
+
"step": 25
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.02,
|
| 36 |
+
"grad_norm": 6.4887261390686035,
|
| 37 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 38 |
+
"loss": 0.5405,
|
| 39 |
+
"step": 30
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.06,
|
| 43 |
+
"grad_norm": 4.2266340255737305,
|
| 44 |
+
"learning_rate": 4.62962962962963e-05,
|
| 45 |
+
"loss": 0.5519,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.1,
|
| 50 |
+
"grad_norm": 38.938472747802734,
|
| 51 |
+
"learning_rate": 4.3981481481481486e-05,
|
| 52 |
+
"loss": 0.6436,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.1,
|
| 57 |
+
"eval_accuracy": 0.5058448459086079,
|
| 58 |
+
"eval_loss": 0.9187208414077759,
|
| 59 |
+
"eval_runtime": 186.1328,
|
| 60 |
+
"eval_samples_per_second": 5.056,
|
| 61 |
+
"eval_steps_per_second": 0.317,
|
| 62 |
+
"step": 50
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.04,
|
| 66 |
+
"grad_norm": 5.287866115570068,
|
| 67 |
+
"learning_rate": 4.166666666666667e-05,
|
| 68 |
+
"loss": 0.4336,
|
| 69 |
+
"step": 60
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 2.08,
|
| 73 |
+
"grad_norm": 8.11617374420166,
|
| 74 |
+
"learning_rate": 3.935185185185186e-05,
|
| 75 |
+
"loss": 0.5464,
|
| 76 |
+
"step": 70
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.1,
|
| 80 |
+
"eval_accuracy": 0.5313496280552603,
|
| 81 |
+
"eval_loss": 0.8164978623390198,
|
| 82 |
+
"eval_runtime": 165.1643,
|
| 83 |
+
"eval_samples_per_second": 5.697,
|
| 84 |
+
"eval_steps_per_second": 0.357,
|
| 85 |
+
"step": 75
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 3.02,
|
| 89 |
+
"grad_norm": 9.028487205505371,
|
| 90 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 91 |
+
"loss": 0.5459,
|
| 92 |
+
"step": 80
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 3.06,
|
| 96 |
+
"grad_norm": 4.760679721832275,
|
| 97 |
+
"learning_rate": 3.472222222222222e-05,
|
| 98 |
+
"loss": 0.4503,
|
| 99 |
+
"step": 90
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 3.1,
|
| 103 |
+
"grad_norm": 13.96630573272705,
|
| 104 |
+
"learning_rate": 3.240740740740741e-05,
|
| 105 |
+
"loss": 0.4147,
|
| 106 |
+
"step": 100
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 3.1,
|
| 110 |
+
"eval_accuracy": 0.5228480340063762,
|
| 111 |
+
"eval_loss": 0.9949731230735779,
|
| 112 |
+
"eval_runtime": 166.6223,
|
| 113 |
+
"eval_samples_per_second": 5.648,
|
| 114 |
+
"eval_steps_per_second": 0.354,
|
| 115 |
+
"step": 100
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 4.04,
|
| 119 |
+
"grad_norm": 3.802431344985962,
|
| 120 |
+
"learning_rate": 3.0092592592592593e-05,
|
| 121 |
+
"loss": 0.3409,
|
| 122 |
+
"step": 110
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 4.08,
|
| 126 |
+
"grad_norm": 9.562992095947266,
|
| 127 |
+
"learning_rate": 2.777777777777778e-05,
|
| 128 |
+
"loss": 0.4483,
|
| 129 |
+
"step": 120
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 4.1,
|
| 133 |
+
"eval_accuracy": 0.5079702444208289,
|
| 134 |
+
"eval_loss": 1.064165711402893,
|
| 135 |
+
"eval_runtime": 194.4496,
|
| 136 |
+
"eval_samples_per_second": 4.839,
|
| 137 |
+
"eval_steps_per_second": 0.303,
|
| 138 |
+
"step": 125
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 5.02,
|
| 142 |
+
"grad_norm": 8.446785926818848,
|
| 143 |
+
"learning_rate": 2.5462962962962965e-05,
|
| 144 |
+
"loss": 0.6922,
|
| 145 |
+
"step": 130
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 5.06,
|
| 149 |
+
"grad_norm": 3.8589024543762207,
|
| 150 |
+
"learning_rate": 2.314814814814815e-05,
|
| 151 |
+
"loss": 0.3718,
|
| 152 |
+
"step": 140
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 5.1,
|
| 156 |
+
"grad_norm": 5.9254536628723145,
|
| 157 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 158 |
+
"loss": 0.3565,
|
| 159 |
+
"step": 150
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 5.1,
|
| 163 |
+
"eval_accuracy": 0.5143464399574921,
|
| 164 |
+
"eval_loss": 0.8610547780990601,
|
| 165 |
+
"eval_runtime": 184.6388,
|
| 166 |
+
"eval_samples_per_second": 5.096,
|
| 167 |
+
"eval_steps_per_second": 0.32,
|
| 168 |
+
"step": 150
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.04,
|
| 172 |
+
"grad_norm": 5.898458003997803,
|
| 173 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 174 |
+
"loss": 0.3232,
|
| 175 |
+
"step": 160
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 6.08,
|
| 179 |
+
"grad_norm": 3.0727579593658447,
|
| 180 |
+
"learning_rate": 1.6203703703703704e-05,
|
| 181 |
+
"loss": 0.3114,
|
| 182 |
+
"step": 170
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.1,
|
| 186 |
+
"eval_accuracy": 0.5185972369819342,
|
| 187 |
+
"eval_loss": 0.9594563245773315,
|
| 188 |
+
"eval_runtime": 185.3195,
|
| 189 |
+
"eval_samples_per_second": 5.078,
|
| 190 |
+
"eval_steps_per_second": 0.318,
|
| 191 |
+
"step": 175
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 7.02,
|
| 195 |
+
"grad_norm": 1.4301713705062866,
|
| 196 |
+
"learning_rate": 1.388888888888889e-05,
|
| 197 |
+
"loss": 0.253,
|
| 198 |
+
"step": 180
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 7.06,
|
| 202 |
+
"grad_norm": 6.7790045738220215,
|
| 203 |
+
"learning_rate": 1.1574074074074075e-05,
|
| 204 |
+
"loss": 0.3144,
|
| 205 |
+
"step": 190
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.1,
|
| 209 |
+
"grad_norm": 26.213571548461914,
|
| 210 |
+
"learning_rate": 9.259259259259259e-06,
|
| 211 |
+
"loss": 0.3074,
|
| 212 |
+
"step": 200
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.1,
|
| 216 |
+
"eval_accuracy": 0.5196599362380446,
|
| 217 |
+
"eval_loss": 1.0241553783416748,
|
| 218 |
+
"eval_runtime": 191.958,
|
| 219 |
+
"eval_samples_per_second": 4.902,
|
| 220 |
+
"eval_steps_per_second": 0.307,
|
| 221 |
+
"step": 200
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 8.04,
|
| 225 |
+
"grad_norm": 3.681013822555542,
|
| 226 |
+
"learning_rate": 6.944444444444445e-06,
|
| 227 |
+
"loss": 0.2552,
|
| 228 |
+
"step": 210
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 8.08,
|
| 232 |
+
"grad_norm": 6.130499839782715,
|
| 233 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 234 |
+
"loss": 0.2888,
|
| 235 |
+
"step": 220
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 8.1,
|
| 239 |
+
"eval_accuracy": 0.5069075451647184,
|
| 240 |
+
"eval_loss": 1.1212139129638672,
|
| 241 |
+
"eval_runtime": 182.5202,
|
| 242 |
+
"eval_samples_per_second": 5.156,
|
| 243 |
+
"eval_steps_per_second": 0.323,
|
| 244 |
+
"step": 225
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 9.02,
|
| 248 |
+
"grad_norm": 4.048097133636475,
|
| 249 |
+
"learning_rate": 2.3148148148148148e-06,
|
| 250 |
+
"loss": 0.2812,
|
| 251 |
+
"step": 230
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 9.06,
|
| 255 |
+
"grad_norm": 2.3711907863616943,
|
| 256 |
+
"learning_rate": 0.0,
|
| 257 |
+
"loss": 0.2609,
|
| 258 |
+
"step": 240
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 9.06,
|
| 262 |
+
"eval_accuracy": 0.5111583421891605,
|
| 263 |
+
"eval_loss": 1.111896276473999,
|
| 264 |
+
"eval_runtime": 168.1115,
|
| 265 |
+
"eval_samples_per_second": 5.597,
|
| 266 |
+
"eval_steps_per_second": 0.351,
|
| 267 |
+
"step": 240
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 9.06,
|
| 271 |
+
"step": 240,
|
| 272 |
+
"total_flos": 3.246125779066061e+18,
|
| 273 |
+
"train_loss": 0.43130043546358743,
|
| 274 |
+
"train_runtime": 2651.285,
|
| 275 |
+
"train_samples_per_second": 1.448,
|
| 276 |
+
"train_steps_per_second": 0.091
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 9.06,
|
| 280 |
+
"eval_accuracy": 0.5011600928074246,
|
| 281 |
+
"eval_loss": 0.8047753572463989,
|
| 282 |
+
"eval_runtime": 160.408,
|
| 283 |
+
"eval_samples_per_second": 5.374,
|
| 284 |
+
"eval_steps_per_second": 0.337,
|
| 285 |
+
"step": 240
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 9.06,
|
| 289 |
+
"eval_accuracy": 0.5011600928074246,
|
| 290 |
+
"eval_loss": 0.8047754168510437,
|
| 291 |
+
"eval_runtime": 160.5206,
|
| 292 |
+
"eval_samples_per_second": 5.37,
|
| 293 |
+
"eval_steps_per_second": 0.336,
|
| 294 |
+
"step": 240
|
| 295 |
+
}
|
| 296 |
+
],
|
| 297 |
+
"logging_steps": 10,
|
| 298 |
+
"max_steps": 240,
|
| 299 |
+
"num_input_tokens_seen": 0,
|
| 300 |
+
"num_train_epochs": 9223372036854775807,
|
| 301 |
+
"save_steps": 500,
|
| 302 |
+
"total_flos": 3.246125779066061e+18,
|
| 303 |
+
"train_batch_size": 16,
|
| 304 |
+
"trial_name": null,
|
| 305 |
+
"trial_params": null
|
| 306 |
+
}
|