End of training
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- train_results.txt +6 -0
- trainer_state.json +274 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 2.181800748630912,
|
| 4 |
+
"train_runtime": 3900.3613,
|
| 5 |
+
"train_samples": 42696,
|
| 6 |
+
"train_samples_per_second": 54.733,
|
| 7 |
+
"train_steps_per_second": 0.856
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 2.181800748630912,
|
| 4 |
+
"train_runtime": 3900.3613,
|
| 5 |
+
"train_samples": 42696,
|
| 6 |
+
"train_samples_per_second": 54.733,
|
| 7 |
+
"train_steps_per_second": 0.856
|
| 8 |
+
}
|
train_results.txt
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
epoch = 5.0
|
| 2 |
+
train_loss = 2.181800748630912
|
| 3 |
+
train_runtime = 3900.3613
|
| 4 |
+
train_samples = 42696
|
| 5 |
+
train_samples_per_second = 54.733
|
| 6 |
+
train_steps_per_second = 0.856
|
trainer_state.json
ADDED
|
@@ -0,0 +1,274 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 2.6812784671783447,
|
| 3 |
+
"best_model_checkpoint": "./models/final_bart/checkpoint-2000",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 3340,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.15,
|
| 12 |
+
"learning_rate": 8.982035928143711e-06,
|
| 13 |
+
"loss": 5.6146,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.3,
|
| 18 |
+
"learning_rate": 1.7964071856287423e-05,
|
| 19 |
+
"loss": 2.8024,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.45,
|
| 24 |
+
"learning_rate": 2.6946107784431136e-05,
|
| 25 |
+
"loss": 2.5446,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.6,
|
| 30 |
+
"learning_rate": 2.934131736526946e-05,
|
| 31 |
+
"loss": 2.4198,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.75,
|
| 36 |
+
"learning_rate": 2.8343313373253494e-05,
|
| 37 |
+
"loss": 2.3636,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.9,
|
| 42 |
+
"learning_rate": 2.7345309381237524e-05,
|
| 43 |
+
"loss": 2.3251,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.05,
|
| 48 |
+
"learning_rate": 2.6347305389221558e-05,
|
| 49 |
+
"loss": 2.2522,
|
| 50 |
+
"step": 700
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.2,
|
| 54 |
+
"learning_rate": 2.534930139720559e-05,
|
| 55 |
+
"loss": 2.1959,
|
| 56 |
+
"step": 800
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.35,
|
| 60 |
+
"learning_rate": 2.4351297405189622e-05,
|
| 61 |
+
"loss": 2.1588,
|
| 62 |
+
"step": 900
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.5,
|
| 66 |
+
"learning_rate": 2.3353293413173656e-05,
|
| 67 |
+
"loss": 2.1542,
|
| 68 |
+
"step": 1000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.5,
|
| 72 |
+
"eval_bleu1": 27.9938,
|
| 73 |
+
"eval_bleu2": 15.5354,
|
| 74 |
+
"eval_bleu3": 8.2494,
|
| 75 |
+
"eval_bleu4": 4.42,
|
| 76 |
+
"eval_gen_len": 50.08,
|
| 77 |
+
"eval_loss": 2.749131202697754,
|
| 78 |
+
"eval_rdass": 0.6093000173568726,
|
| 79 |
+
"eval_rouge1": 33.5554,
|
| 80 |
+
"eval_rouge2": 11.2371,
|
| 81 |
+
"eval_rougeL": 22.006,
|
| 82 |
+
"eval_runtime": 24.7242,
|
| 83 |
+
"eval_samples_per_second": 4.045,
|
| 84 |
+
"eval_steps_per_second": 0.04,
|
| 85 |
+
"step": 1000
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 1.65,
|
| 89 |
+
"learning_rate": 2.2355289421157686e-05,
|
| 90 |
+
"loss": 2.1473,
|
| 91 |
+
"step": 1100
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 1.8,
|
| 95 |
+
"learning_rate": 2.135728542914172e-05,
|
| 96 |
+
"loss": 2.1459,
|
| 97 |
+
"step": 1200
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 1.95,
|
| 101 |
+
"learning_rate": 2.0359281437125747e-05,
|
| 102 |
+
"loss": 2.1324,
|
| 103 |
+
"step": 1300
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 2.1,
|
| 107 |
+
"learning_rate": 1.936127744510978e-05,
|
| 108 |
+
"loss": 2.0676,
|
| 109 |
+
"step": 1400
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.25,
|
| 113 |
+
"learning_rate": 1.836327345309381e-05,
|
| 114 |
+
"loss": 2.0206,
|
| 115 |
+
"step": 1500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 2.4,
|
| 119 |
+
"learning_rate": 1.7365269461077845e-05,
|
| 120 |
+
"loss": 2.0198,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 2.54,
|
| 125 |
+
"learning_rate": 1.6367265469061875e-05,
|
| 126 |
+
"loss": 2.0177,
|
| 127 |
+
"step": 1700
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 2.69,
|
| 131 |
+
"learning_rate": 1.536926147704591e-05,
|
| 132 |
+
"loss": 2.0143,
|
| 133 |
+
"step": 1800
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 2.84,
|
| 137 |
+
"learning_rate": 1.437125748502994e-05,
|
| 138 |
+
"loss": 2.0075,
|
| 139 |
+
"step": 1900
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.99,
|
| 143 |
+
"learning_rate": 1.3373253493013973e-05,
|
| 144 |
+
"loss": 2.0071,
|
| 145 |
+
"step": 2000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 2.99,
|
| 149 |
+
"eval_bleu1": 29.6866,
|
| 150 |
+
"eval_bleu2": 17.1396,
|
| 151 |
+
"eval_bleu3": 9.7016,
|
| 152 |
+
"eval_bleu4": 5.3559,
|
| 153 |
+
"eval_gen_len": 54.04,
|
| 154 |
+
"eval_loss": 2.6812784671783447,
|
| 155 |
+
"eval_rdass": 0.6154999732971191,
|
| 156 |
+
"eval_rouge1": 35.0501,
|
| 157 |
+
"eval_rouge2": 12.2759,
|
| 158 |
+
"eval_rougeL": 22.6669,
|
| 159 |
+
"eval_runtime": 20.0572,
|
| 160 |
+
"eval_samples_per_second": 4.986,
|
| 161 |
+
"eval_steps_per_second": 0.05,
|
| 162 |
+
"step": 2000
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.14,
|
| 166 |
+
"learning_rate": 1.2375249500998005e-05,
|
| 167 |
+
"loss": 1.9318,
|
| 168 |
+
"step": 2100
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 3.29,
|
| 172 |
+
"learning_rate": 1.1377245508982035e-05,
|
| 173 |
+
"loss": 1.9389,
|
| 174 |
+
"step": 2200
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 3.44,
|
| 178 |
+
"learning_rate": 1.0379241516966067e-05,
|
| 179 |
+
"loss": 1.9214,
|
| 180 |
+
"step": 2300
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 3.59,
|
| 184 |
+
"learning_rate": 9.3812375249501e-06,
|
| 185 |
+
"loss": 1.9144,
|
| 186 |
+
"step": 2400
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 3.74,
|
| 190 |
+
"learning_rate": 8.383233532934131e-06,
|
| 191 |
+
"loss": 1.9404,
|
| 192 |
+
"step": 2500
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 3.89,
|
| 196 |
+
"learning_rate": 7.385229540918164e-06,
|
| 197 |
+
"loss": 1.9189,
|
| 198 |
+
"step": 2600
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 4.04,
|
| 202 |
+
"learning_rate": 6.3872255489021955e-06,
|
| 203 |
+
"loss": 1.9155,
|
| 204 |
+
"step": 2700
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 4.19,
|
| 208 |
+
"learning_rate": 5.3892215568862275e-06,
|
| 209 |
+
"loss": 1.874,
|
| 210 |
+
"step": 2800
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 4.34,
|
| 214 |
+
"learning_rate": 4.39121756487026e-06,
|
| 215 |
+
"loss": 1.8951,
|
| 216 |
+
"step": 2900
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 4.49,
|
| 220 |
+
"learning_rate": 3.3932135728542917e-06,
|
| 221 |
+
"loss": 1.8694,
|
| 222 |
+
"step": 3000
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 4.49,
|
| 226 |
+
"eval_bleu1": 30.5261,
|
| 227 |
+
"eval_bleu2": 17.6264,
|
| 228 |
+
"eval_bleu3": 10.3974,
|
| 229 |
+
"eval_bleu4": 5.4348,
|
| 230 |
+
"eval_gen_len": 53.47,
|
| 231 |
+
"eval_loss": 2.684814453125,
|
| 232 |
+
"eval_rdass": 0.6248000264167786,
|
| 233 |
+
"eval_rouge1": 35.7722,
|
| 234 |
+
"eval_rouge2": 12.5127,
|
| 235 |
+
"eval_rougeL": 23.3002,
|
| 236 |
+
"eval_runtime": 17.6216,
|
| 237 |
+
"eval_samples_per_second": 5.675,
|
| 238 |
+
"eval_steps_per_second": 0.057,
|
| 239 |
+
"step": 3000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 4.64,
|
| 243 |
+
"learning_rate": 2.3952095808383233e-06,
|
| 244 |
+
"loss": 1.8708,
|
| 245 |
+
"step": 3100
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 4.79,
|
| 249 |
+
"learning_rate": 1.3972055888223554e-06,
|
| 250 |
+
"loss": 1.8628,
|
| 251 |
+
"step": 3200
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 4.94,
|
| 255 |
+
"learning_rate": 3.992015968063872e-07,
|
| 256 |
+
"loss": 1.8613,
|
| 257 |
+
"step": 3300
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 5.0,
|
| 261 |
+
"step": 3340,
|
| 262 |
+
"total_flos": 5.5473275160576e+16,
|
| 263 |
+
"train_loss": 2.181800748630912,
|
| 264 |
+
"train_runtime": 3900.3613,
|
| 265 |
+
"train_samples_per_second": 54.733,
|
| 266 |
+
"train_steps_per_second": 0.856
|
| 267 |
+
}
|
| 268 |
+
],
|
| 269 |
+
"max_steps": 3340,
|
| 270 |
+
"num_train_epochs": 5,
|
| 271 |
+
"total_flos": 5.5473275160576e+16,
|
| 272 |
+
"trial_name": null,
|
| 273 |
+
"trial_params": null
|
| 274 |
+
}
|