End of training
Browse files- trainer_state.json +282 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,282 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9625,
|
| 3 |
+
"best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-asl/checkpoint-90",
|
| 4 |
+
"epoch": 3.25,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 360,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 1.388888888888889e-05,
|
| 14 |
+
"loss": 2.0901,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.06,
|
| 19 |
+
"learning_rate": 2.777777777777778e-05,
|
| 20 |
+
"loss": 1.7266,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.08,
|
| 25 |
+
"learning_rate": 4.166666666666667e-05,
|
| 26 |
+
"loss": 1.4769,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.11,
|
| 31 |
+
"learning_rate": 4.938271604938271e-05,
|
| 32 |
+
"loss": 1.0936,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.14,
|
| 37 |
+
"learning_rate": 4.783950617283951e-05,
|
| 38 |
+
"loss": 0.824,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.17,
|
| 43 |
+
"learning_rate": 4.62962962962963e-05,
|
| 44 |
+
"loss": 0.8398,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.19,
|
| 49 |
+
"learning_rate": 4.4753086419753084e-05,
|
| 50 |
+
"loss": 0.435,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.22,
|
| 55 |
+
"learning_rate": 4.3209876543209875e-05,
|
| 56 |
+
"loss": 0.6051,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.25,
|
| 61 |
+
"learning_rate": 4.166666666666667e-05,
|
| 62 |
+
"loss": 0.3325,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.25,
|
| 67 |
+
"eval_accuracy": 0.9625,
|
| 68 |
+
"eval_loss": 0.2076590359210968,
|
| 69 |
+
"eval_runtime": 78.5536,
|
| 70 |
+
"eval_samples_per_second": 1.018,
|
| 71 |
+
"eval_steps_per_second": 0.127,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.03,
|
| 76 |
+
"learning_rate": 4.012345679012346e-05,
|
| 77 |
+
"loss": 0.3724,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.06,
|
| 82 |
+
"learning_rate": 3.8580246913580246e-05,
|
| 83 |
+
"loss": 0.3341,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.08,
|
| 88 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 89 |
+
"loss": 0.2816,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.11,
|
| 94 |
+
"learning_rate": 3.5493827160493834e-05,
|
| 95 |
+
"loss": 0.393,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.14,
|
| 100 |
+
"learning_rate": 3.395061728395062e-05,
|
| 101 |
+
"loss": 0.1777,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.17,
|
| 106 |
+
"learning_rate": 3.240740740740741e-05,
|
| 107 |
+
"loss": 0.3507,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.19,
|
| 112 |
+
"learning_rate": 3.08641975308642e-05,
|
| 113 |
+
"loss": 0.462,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.22,
|
| 118 |
+
"learning_rate": 2.9320987654320992e-05,
|
| 119 |
+
"loss": 0.3912,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.25,
|
| 124 |
+
"learning_rate": 2.777777777777778e-05,
|
| 125 |
+
"loss": 0.2898,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.25,
|
| 130 |
+
"eval_accuracy": 0.9375,
|
| 131 |
+
"eval_loss": 0.1923440843820572,
|
| 132 |
+
"eval_runtime": 76.3,
|
| 133 |
+
"eval_samples_per_second": 1.048,
|
| 134 |
+
"eval_steps_per_second": 0.131,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.03,
|
| 139 |
+
"learning_rate": 2.623456790123457e-05,
|
| 140 |
+
"loss": 0.2911,
|
| 141 |
+
"step": 190
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 2.06,
|
| 145 |
+
"learning_rate": 2.4691358024691357e-05,
|
| 146 |
+
"loss": 0.3739,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.08,
|
| 151 |
+
"learning_rate": 2.314814814814815e-05,
|
| 152 |
+
"loss": 0.1862,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.11,
|
| 157 |
+
"learning_rate": 2.1604938271604937e-05,
|
| 158 |
+
"loss": 0.1881,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.14,
|
| 163 |
+
"learning_rate": 2.006172839506173e-05,
|
| 164 |
+
"loss": 0.2464,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.17,
|
| 169 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 170 |
+
"loss": 0.2937,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.19,
|
| 175 |
+
"learning_rate": 1.697530864197531e-05,
|
| 176 |
+
"loss": 0.3399,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.22,
|
| 181 |
+
"learning_rate": 1.54320987654321e-05,
|
| 182 |
+
"loss": 0.2323,
|
| 183 |
+
"step": 260
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.25,
|
| 187 |
+
"learning_rate": 1.388888888888889e-05,
|
| 188 |
+
"loss": 0.3358,
|
| 189 |
+
"step": 270
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.25,
|
| 193 |
+
"eval_accuracy": 0.95,
|
| 194 |
+
"eval_loss": 0.11704353988170624,
|
| 195 |
+
"eval_runtime": 78.6154,
|
| 196 |
+
"eval_samples_per_second": 1.018,
|
| 197 |
+
"eval_steps_per_second": 0.127,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 3.03,
|
| 202 |
+
"learning_rate": 1.2345679012345678e-05,
|
| 203 |
+
"loss": 0.2584,
|
| 204 |
+
"step": 280
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.06,
|
| 208 |
+
"learning_rate": 1.0802469135802469e-05,
|
| 209 |
+
"loss": 0.2827,
|
| 210 |
+
"step": 290
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 3.08,
|
| 214 |
+
"learning_rate": 9.259259259259259e-06,
|
| 215 |
+
"loss": 0.1856,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.11,
|
| 220 |
+
"learning_rate": 7.71604938271605e-06,
|
| 221 |
+
"loss": 0.3021,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.14,
|
| 226 |
+
"learning_rate": 6.172839506172839e-06,
|
| 227 |
+
"loss": 0.3106,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.17,
|
| 232 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 233 |
+
"loss": 0.2224,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 3.19,
|
| 238 |
+
"learning_rate": 3.0864197530864196e-06,
|
| 239 |
+
"loss": 0.145,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.22,
|
| 244 |
+
"learning_rate": 1.5432098765432098e-06,
|
| 245 |
+
"loss": 0.174,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.25,
|
| 250 |
+
"learning_rate": 0.0,
|
| 251 |
+
"loss": 0.3354,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 3.25,
|
| 256 |
+
"eval_accuracy": 0.9625,
|
| 257 |
+
"eval_loss": 0.1142113208770752,
|
| 258 |
+
"eval_runtime": 76.7347,
|
| 259 |
+
"eval_samples_per_second": 1.043,
|
| 260 |
+
"eval_steps_per_second": 0.13,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 3.25,
|
| 265 |
+
"step": 360,
|
| 266 |
+
"total_flos": 2.523400261997691e+18,
|
| 267 |
+
"train_loss": 0.4772139224741194,
|
| 268 |
+
"train_runtime": 3683.9124,
|
| 269 |
+
"train_samples_per_second": 0.782,
|
| 270 |
+
"train_steps_per_second": 0.098
|
| 271 |
+
}
|
| 272 |
+
],
|
| 273 |
+
"logging_steps": 10,
|
| 274 |
+
"max_steps": 360,
|
| 275 |
+
"num_input_tokens_seen": 0,
|
| 276 |
+
"num_train_epochs": 9223372036854775807,
|
| 277 |
+
"save_steps": 500,
|
| 278 |
+
"total_flos": 2.523400261997691e+18,
|
| 279 |
+
"train_batch_size": 8,
|
| 280 |
+
"trial_name": null,
|
| 281 |
+
"trial_params": null
|
| 282 |
+
}
|