End of training
Browse files- all_results.json +19 -0
- test_results.json +13 -0
- train_results.json +8 -0
- trainer_state.json +283 -0
all_results.json
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_samples": 1233,
|
| 4 |
+
"test_false_f1": 0.7841191066997518,
|
| 5 |
+
"test_loss": 0.672432005405426,
|
| 6 |
+
"test_macro_f1": 0.6707393317114294,
|
| 7 |
+
"test_micro_f1": 0.7907542579075426,
|
| 8 |
+
"test_mixture_f1": 0.5116279069767442,
|
| 9 |
+
"test_runtime": 40.6849,
|
| 10 |
+
"test_samples_per_second": 30.306,
|
| 11 |
+
"test_steps_per_second": 0.959,
|
| 12 |
+
"test_true_f1": 0.9061976549413736,
|
| 13 |
+
"test_unproven_f1": 0.48101265822784806,
|
| 14 |
+
"train_loss": 0.5249485609128204,
|
| 15 |
+
"train_runtime": 3291.1672,
|
| 16 |
+
"train_samples": 9804,
|
| 17 |
+
"train_samples_per_second": 8.937,
|
| 18 |
+
"train_steps_per_second": 1.118
|
| 19 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_samples": 1233,
|
| 3 |
+
"test_false_f1": 0.7841191066997518,
|
| 4 |
+
"test_loss": 0.672432005405426,
|
| 5 |
+
"test_macro_f1": 0.6707393317114294,
|
| 6 |
+
"test_micro_f1": 0.7907542579075426,
|
| 7 |
+
"test_mixture_f1": 0.5116279069767442,
|
| 8 |
+
"test_runtime": 40.6849,
|
| 9 |
+
"test_samples_per_second": 30.306,
|
| 10 |
+
"test_steps_per_second": 0.959,
|
| 11 |
+
"test_true_f1": 0.9061976549413736,
|
| 12 |
+
"test_unproven_f1": 0.48101265822784806
|
| 13 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 0.5249485609128204,
|
| 4 |
+
"train_runtime": 3291.1672,
|
| 5 |
+
"train_samples": 9804,
|
| 6 |
+
"train_samples_per_second": 8.937,
|
| 7 |
+
"train_steps_per_second": 1.118
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,283 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8179571663920923,
|
| 3 |
+
"best_model_checkpoint": "./bigbird-base-health-fact/checkpoint-2452",
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"global_step": 3678,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.08,
|
| 12 |
+
"learning_rate": 2.6630434782608698e-06,
|
| 13 |
+
"loss": 1.2792,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.16,
|
| 18 |
+
"learning_rate": 5.380434782608695e-06,
|
| 19 |
+
"loss": 1.0546,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.24,
|
| 24 |
+
"learning_rate": 8.097826086956523e-06,
|
| 25 |
+
"loss": 0.8857,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.33,
|
| 30 |
+
"learning_rate": 9.909365558912388e-06,
|
| 31 |
+
"loss": 0.7958,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.41,
|
| 36 |
+
"learning_rate": 9.60725075528701e-06,
|
| 37 |
+
"loss": 0.7246,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.49,
|
| 42 |
+
"learning_rate": 9.305135951661632e-06,
|
| 43 |
+
"loss": 0.7629,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.57,
|
| 48 |
+
"learning_rate": 9.003021148036256e-06,
|
| 49 |
+
"loss": 0.6498,
|
| 50 |
+
"step": 700
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.65,
|
| 54 |
+
"learning_rate": 8.700906344410877e-06,
|
| 55 |
+
"loss": 0.6201,
|
| 56 |
+
"step": 800
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.73,
|
| 60 |
+
"learning_rate": 8.398791540785499e-06,
|
| 61 |
+
"loss": 0.5809,
|
| 62 |
+
"step": 900
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.82,
|
| 66 |
+
"learning_rate": 8.099697885196374e-06,
|
| 67 |
+
"loss": 0.647,
|
| 68 |
+
"step": 1000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.9,
|
| 72 |
+
"learning_rate": 7.797583081570997e-06,
|
| 73 |
+
"loss": 0.5814,
|
| 74 |
+
"step": 1100
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.98,
|
| 78 |
+
"learning_rate": 7.4954682779456205e-06,
|
| 79 |
+
"loss": 0.5563,
|
| 80 |
+
"step": 1200
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.0,
|
| 84 |
+
"eval_false_f1": 0.7926023778071335,
|
| 85 |
+
"eval_loss": 0.5019509196281433,
|
| 86 |
+
"eval_macro_f1": 0.6062122975261963,
|
| 87 |
+
"eval_micro_f1": 0.7948929159802306,
|
| 88 |
+
"eval_mixture_f1": 0.4591194968553459,
|
| 89 |
+
"eval_runtime": 39.3928,
|
| 90 |
+
"eval_samples_per_second": 30.818,
|
| 91 |
+
"eval_steps_per_second": 0.965,
|
| 92 |
+
"eval_true_f1": 0.8986175115207373,
|
| 93 |
+
"eval_unproven_f1": 0.2745098039215686,
|
| 94 |
+
"step": 1226
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.06,
|
| 98 |
+
"learning_rate": 7.193353474320243e-06,
|
| 99 |
+
"loss": 0.5271,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 1.14,
|
| 104 |
+
"learning_rate": 6.891238670694864e-06,
|
| 105 |
+
"loss": 0.4912,
|
| 106 |
+
"step": 1400
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.22,
|
| 110 |
+
"learning_rate": 6.589123867069487e-06,
|
| 111 |
+
"loss": 0.4844,
|
| 112 |
+
"step": 1500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 1.31,
|
| 116 |
+
"learning_rate": 6.287009063444109e-06,
|
| 117 |
+
"loss": 0.4816,
|
| 118 |
+
"step": 1600
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 1.39,
|
| 122 |
+
"learning_rate": 5.984894259818732e-06,
|
| 123 |
+
"loss": 0.462,
|
| 124 |
+
"step": 1700
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.47,
|
| 128 |
+
"learning_rate": 5.682779456193354e-06,
|
| 129 |
+
"loss": 0.4087,
|
| 130 |
+
"step": 1800
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.55,
|
| 134 |
+
"learning_rate": 5.380664652567976e-06,
|
| 135 |
+
"loss": 0.5065,
|
| 136 |
+
"step": 1900
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.63,
|
| 140 |
+
"learning_rate": 5.078549848942599e-06,
|
| 141 |
+
"loss": 0.4313,
|
| 142 |
+
"step": 2000
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.71,
|
| 146 |
+
"learning_rate": 4.776435045317221e-06,
|
| 147 |
+
"loss": 0.5098,
|
| 148 |
+
"step": 2100
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 1.79,
|
| 152 |
+
"learning_rate": 4.4743202416918435e-06,
|
| 153 |
+
"loss": 0.4699,
|
| 154 |
+
"step": 2200
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 1.88,
|
| 158 |
+
"learning_rate": 4.172205438066466e-06,
|
| 159 |
+
"loss": 0.4408,
|
| 160 |
+
"step": 2300
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 1.96,
|
| 164 |
+
"learning_rate": 3.8700906344410875e-06,
|
| 165 |
+
"loss": 0.5048,
|
| 166 |
+
"step": 2400
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 2.0,
|
| 170 |
+
"eval_false_f1": 0.8201811125485123,
|
| 171 |
+
"eval_loss": 0.4968700110912323,
|
| 172 |
+
"eval_macro_f1": 0.684587518040316,
|
| 173 |
+
"eval_micro_f1": 0.8179571663920923,
|
| 174 |
+
"eval_mixture_f1": 0.43416370106761565,
|
| 175 |
+
"eval_runtime": 39.4059,
|
| 176 |
+
"eval_samples_per_second": 30.808,
|
| 177 |
+
"eval_steps_per_second": 0.964,
|
| 178 |
+
"eval_true_f1": 0.9125766871165646,
|
| 179 |
+
"eval_unproven_f1": 0.5714285714285714,
|
| 180 |
+
"step": 2452
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 2.04,
|
| 184 |
+
"learning_rate": 3.5679758308157103e-06,
|
| 185 |
+
"loss": 0.378,
|
| 186 |
+
"step": 2500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.12,
|
| 190 |
+
"learning_rate": 3.2658610271903322e-06,
|
| 191 |
+
"loss": 0.3631,
|
| 192 |
+
"step": 2600
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.2,
|
| 196 |
+
"learning_rate": 2.963746223564955e-06,
|
| 197 |
+
"loss": 0.3475,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 2.28,
|
| 202 |
+
"learning_rate": 2.661631419939577e-06,
|
| 203 |
+
"loss": 0.3283,
|
| 204 |
+
"step": 2800
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.37,
|
| 208 |
+
"learning_rate": 2.3595166163142e-06,
|
| 209 |
+
"loss": 0.317,
|
| 210 |
+
"step": 2900
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.45,
|
| 214 |
+
"learning_rate": 2.0604229607250755e-06,
|
| 215 |
+
"loss": 0.3541,
|
| 216 |
+
"step": 3000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.53,
|
| 220 |
+
"learning_rate": 1.758308157099698e-06,
|
| 221 |
+
"loss": 0.3818,
|
| 222 |
+
"step": 3100
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.61,
|
| 226 |
+
"learning_rate": 1.4561933534743203e-06,
|
| 227 |
+
"loss": 0.3467,
|
| 228 |
+
"step": 3200
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 2.69,
|
| 232 |
+
"learning_rate": 1.1540785498489427e-06,
|
| 233 |
+
"loss": 0.3048,
|
| 234 |
+
"step": 3300
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 2.77,
|
| 238 |
+
"learning_rate": 8.51963746223565e-07,
|
| 239 |
+
"loss": 0.3334,
|
| 240 |
+
"step": 3400
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.85,
|
| 244 |
+
"learning_rate": 5.498489425981874e-07,
|
| 245 |
+
"loss": 0.3855,
|
| 246 |
+
"step": 3500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 2.94,
|
| 250 |
+
"learning_rate": 2.477341389728097e-07,
|
| 251 |
+
"loss": 0.3454,
|
| 252 |
+
"step": 3600
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 3.0,
|
| 256 |
+
"eval_false_f1": 0.811443433029909,
|
| 257 |
+
"eval_loss": 0.5863622426986694,
|
| 258 |
+
"eval_macro_f1": 0.6874160790583576,
|
| 259 |
+
"eval_micro_f1": 0.8130148270181219,
|
| 260 |
+
"eval_mixture_f1": 0.4556962025316456,
|
| 261 |
+
"eval_runtime": 39.6082,
|
| 262 |
+
"eval_samples_per_second": 30.65,
|
| 263 |
+
"eval_steps_per_second": 0.959,
|
| 264 |
+
"eval_true_f1": 0.9153605015673981,
|
| 265 |
+
"eval_unproven_f1": 0.5671641791044776,
|
| 266 |
+
"step": 3678
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 3.0,
|
| 270 |
+
"step": 3678,
|
| 271 |
+
"total_flos": 2.106512041918464e+16,
|
| 272 |
+
"train_loss": 0.5249485609128204,
|
| 273 |
+
"train_runtime": 3291.1672,
|
| 274 |
+
"train_samples_per_second": 8.937,
|
| 275 |
+
"train_steps_per_second": 1.118
|
| 276 |
+
}
|
| 277 |
+
],
|
| 278 |
+
"max_steps": 3678,
|
| 279 |
+
"num_train_epochs": 3,
|
| 280 |
+
"total_flos": 2.106512041918464e+16,
|
| 281 |
+
"trial_name": null,
|
| 282 |
+
"trial_params": null
|
| 283 |
+
}
|