End of training
Browse files- all_results.json +15 -0
- eval_nbest_predictions.json +2 -2
- eval_predictions.json +0 -0
- eval_results.json +9 -0
- runs/May16_04-20-09_ml-machine/events.out.tfevents.1715839108.ml-machine.1495.1 +3 -0
- train_results.json +9 -0
- trainer_state.json +280 -0
all_results.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 62.21875,
|
| 4 |
+
"eval_f1": 76.43196644229354,
|
| 5 |
+
"eval_runtime": 54.6387,
|
| 6 |
+
"eval_samples": 4096,
|
| 7 |
+
"eval_samples_per_second": 74.965,
|
| 8 |
+
"eval_steps_per_second": 2.69,
|
| 9 |
+
"total_flos": 3.788022162461184e+16,
|
| 10 |
+
"train_loss": 0.6408887451878504,
|
| 11 |
+
"train_runtime": 5720.422,
|
| 12 |
+
"train_samples": 19329,
|
| 13 |
+
"train_samples_per_second": 33.789,
|
| 14 |
+
"train_steps_per_second": 1.208
|
| 15 |
+
}
|
eval_nbest_predictions.json
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8eda27adddb52b1eb8d43ddc70eb8341fa3488b7d09f88ec45fd0ddfe8fe6a9b
|
| 3 |
+
size 23157655
|
eval_predictions.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_exact_match": 62.21875,
|
| 4 |
+
"eval_f1": 76.43196644229354,
|
| 5 |
+
"eval_runtime": 54.6387,
|
| 6 |
+
"eval_samples": 4096,
|
| 7 |
+
"eval_samples_per_second": 74.965,
|
| 8 |
+
"eval_steps_per_second": 2.69
|
| 9 |
+
}
|
runs/May16_04-20-09_ml-machine/events.out.tfevents.1715839108.ml-machine.1495.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d07c5d38c4f6281cabfe598ab4932ed67da9fa8f4d4550dd3f6b0adffc97b925
|
| 3 |
+
size 412
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 3.788022162461184e+16,
|
| 4 |
+
"train_loss": 0.6408887451878504,
|
| 5 |
+
"train_runtime": 5720.422,
|
| 6 |
+
"train_samples": 19329,
|
| 7 |
+
"train_samples_per_second": 33.789,
|
| 8 |
+
"train_steps_per_second": 1.208
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,280 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 76.43196644229354,
|
| 3 |
+
"best_model_checkpoint": "/root/turkic_qa/en_kaz_models/en_kaz_mdeberta_base_squad_model/checkpoint-2764",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 6910,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"step": 691,
|
| 14 |
+
"train_exact_match": 63.03696303696304,
|
| 15 |
+
"train_f1": 78.13915659047835,
|
| 16 |
+
"train_runtime": 17.4035,
|
| 17 |
+
"train_samples_per_second": 74.468,
|
| 18 |
+
"train_steps_per_second": 2.701
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.0,
|
| 22 |
+
"grad_norm": 40.22502899169922,
|
| 23 |
+
"learning_rate": 5e-06,
|
| 24 |
+
"loss": 1.2642,
|
| 25 |
+
"step": 691
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.0,
|
| 29 |
+
"eval_exact_match": 58.5625,
|
| 30 |
+
"eval_f1": 74.11433052314848,
|
| 31 |
+
"eval_runtime": 54.8873,
|
| 32 |
+
"eval_samples_per_second": 74.626,
|
| 33 |
+
"eval_steps_per_second": 2.678,
|
| 34 |
+
"step": 691
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.0,
|
| 38 |
+
"step": 1382,
|
| 39 |
+
"train_exact_match": 67.33266733266733,
|
| 40 |
+
"train_f1": 80.90558473928773,
|
| 41 |
+
"train_runtime": 17.2817,
|
| 42 |
+
"train_samples_per_second": 75.745,
|
| 43 |
+
"train_steps_per_second": 2.72
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 2.0,
|
| 47 |
+
"grad_norm": 28.721803665161133,
|
| 48 |
+
"learning_rate": 1e-05,
|
| 49 |
+
"loss": 1.0309,
|
| 50 |
+
"step": 1382
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 2.0,
|
| 54 |
+
"eval_exact_match": 60.84375,
|
| 55 |
+
"eval_f1": 75.39680273889411,
|
| 56 |
+
"eval_runtime": 54.2212,
|
| 57 |
+
"eval_samples_per_second": 75.542,
|
| 58 |
+
"eval_steps_per_second": 2.711,
|
| 59 |
+
"step": 1382
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 3.0,
|
| 63 |
+
"step": 2073,
|
| 64 |
+
"train_exact_match": 72.22777222777223,
|
| 65 |
+
"train_f1": 86.54756181813792,
|
| 66 |
+
"train_runtime": 16.9462,
|
| 67 |
+
"train_samples_per_second": 74.943,
|
| 68 |
+
"train_steps_per_second": 2.714
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 3.0,
|
| 72 |
+
"grad_norm": 25.19014549255371,
|
| 73 |
+
"learning_rate": 8.750000000000001e-06,
|
| 74 |
+
"loss": 0.8566,
|
| 75 |
+
"step": 2073
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 3.0,
|
| 79 |
+
"eval_exact_match": 62.1875,
|
| 80 |
+
"eval_f1": 76.29400510573826,
|
| 81 |
+
"eval_runtime": 54.2582,
|
| 82 |
+
"eval_samples_per_second": 75.491,
|
| 83 |
+
"eval_steps_per_second": 2.709,
|
| 84 |
+
"step": 2073
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 4.0,
|
| 88 |
+
"step": 2764,
|
| 89 |
+
"train_exact_match": 75.52447552447552,
|
| 90 |
+
"train_f1": 88.34455431353304,
|
| 91 |
+
"train_runtime": 17.9414,
|
| 92 |
+
"train_samples_per_second": 75.078,
|
| 93 |
+
"train_steps_per_second": 2.731
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 4.0,
|
| 97 |
+
"grad_norm": 31.350345611572266,
|
| 98 |
+
"learning_rate": 7.500000000000001e-06,
|
| 99 |
+
"loss": 0.7007,
|
| 100 |
+
"step": 2764
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 4.0,
|
| 104 |
+
"eval_exact_match": 62.21875,
|
| 105 |
+
"eval_f1": 76.43196644229354,
|
| 106 |
+
"eval_runtime": 54.0444,
|
| 107 |
+
"eval_samples_per_second": 75.79,
|
| 108 |
+
"eval_steps_per_second": 2.72,
|
| 109 |
+
"step": 2764
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 5.0,
|
| 113 |
+
"step": 3455,
|
| 114 |
+
"train_exact_match": 80.01998001998003,
|
| 115 |
+
"train_f1": 91.67585411210774,
|
| 116 |
+
"train_runtime": 17.0774,
|
| 117 |
+
"train_samples_per_second": 75.246,
|
| 118 |
+
"train_steps_per_second": 2.694
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 5.0,
|
| 122 |
+
"grad_norm": 41.42879867553711,
|
| 123 |
+
"learning_rate": 6.25e-06,
|
| 124 |
+
"loss": 0.5836,
|
| 125 |
+
"step": 3455
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 5.0,
|
| 129 |
+
"eval_exact_match": 62.0,
|
| 130 |
+
"eval_f1": 75.89595740381746,
|
| 131 |
+
"eval_runtime": 54.8769,
|
| 132 |
+
"eval_samples_per_second": 74.64,
|
| 133 |
+
"eval_steps_per_second": 2.679,
|
| 134 |
+
"step": 3455
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 6.0,
|
| 138 |
+
"step": 4146,
|
| 139 |
+
"train_exact_match": 81.71828171828172,
|
| 140 |
+
"train_f1": 92.46355571003183,
|
| 141 |
+
"train_runtime": 16.6375,
|
| 142 |
+
"train_samples_per_second": 75.252,
|
| 143 |
+
"train_steps_per_second": 2.705
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 6.0,
|
| 147 |
+
"grad_norm": 20.05802345275879,
|
| 148 |
+
"learning_rate": 5e-06,
|
| 149 |
+
"loss": 0.494,
|
| 150 |
+
"step": 4146
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 6.0,
|
| 154 |
+
"eval_exact_match": 62.46875,
|
| 155 |
+
"eval_f1": 75.95799025269356,
|
| 156 |
+
"eval_runtime": 54.1924,
|
| 157 |
+
"eval_samples_per_second": 75.583,
|
| 158 |
+
"eval_steps_per_second": 2.713,
|
| 159 |
+
"step": 4146
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 7.0,
|
| 163 |
+
"step": 4837,
|
| 164 |
+
"train_exact_match": 82.11788211788212,
|
| 165 |
+
"train_f1": 93.12695306139514,
|
| 166 |
+
"train_runtime": 17.1308,
|
| 167 |
+
"train_samples_per_second": 74.953,
|
| 168 |
+
"train_steps_per_second": 2.685
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 7.0,
|
| 172 |
+
"grad_norm": 36.17577362060547,
|
| 173 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 174 |
+
"loss": 0.4281,
|
| 175 |
+
"step": 4837
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 7.0,
|
| 179 |
+
"eval_exact_match": 62.09375,
|
| 180 |
+
"eval_f1": 75.87457028727934,
|
| 181 |
+
"eval_runtime": 54.4468,
|
| 182 |
+
"eval_samples_per_second": 75.229,
|
| 183 |
+
"eval_steps_per_second": 2.7,
|
| 184 |
+
"step": 4837
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 8.0,
|
| 188 |
+
"step": 5528,
|
| 189 |
+
"train_exact_match": 86.01398601398601,
|
| 190 |
+
"train_f1": 94.42525886876679,
|
| 191 |
+
"train_runtime": 17.0114,
|
| 192 |
+
"train_samples_per_second": 75.479,
|
| 193 |
+
"train_steps_per_second": 2.704
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.0,
|
| 197 |
+
"grad_norm": 12.529890060424805,
|
| 198 |
+
"learning_rate": 2.5e-06,
|
| 199 |
+
"loss": 0.3806,
|
| 200 |
+
"step": 5528
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 8.0,
|
| 204 |
+
"eval_exact_match": 62.0625,
|
| 205 |
+
"eval_f1": 75.78487868224668,
|
| 206 |
+
"eval_runtime": 54.4766,
|
| 207 |
+
"eval_samples_per_second": 75.188,
|
| 208 |
+
"eval_steps_per_second": 2.698,
|
| 209 |
+
"step": 5528
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.0,
|
| 213 |
+
"step": 6219,
|
| 214 |
+
"train_exact_match": 86.11388611388611,
|
| 215 |
+
"train_f1": 94.8958657040008,
|
| 216 |
+
"train_runtime": 17.878,
|
| 217 |
+
"train_samples_per_second": 72.1,
|
| 218 |
+
"train_steps_per_second": 2.629
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 9.0,
|
| 222 |
+
"grad_norm": 37.82954788208008,
|
| 223 |
+
"learning_rate": 1.25e-06,
|
| 224 |
+
"loss": 0.3457,
|
| 225 |
+
"step": 6219
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 9.0,
|
| 229 |
+
"eval_exact_match": 62.03125,
|
| 230 |
+
"eval_f1": 75.65181251557605,
|
| 231 |
+
"eval_runtime": 54.6102,
|
| 232 |
+
"eval_samples_per_second": 75.004,
|
| 233 |
+
"eval_steps_per_second": 2.692,
|
| 234 |
+
"step": 6219
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 10.0,
|
| 238 |
+
"step": 6910,
|
| 239 |
+
"train_exact_match": 87.81218781218782,
|
| 240 |
+
"train_f1": 95.32192987518101,
|
| 241 |
+
"train_runtime": 16.6546,
|
| 242 |
+
"train_samples_per_second": 74.814,
|
| 243 |
+
"train_steps_per_second": 2.702
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 10.0,
|
| 247 |
+
"grad_norm": 8.968611717224121,
|
| 248 |
+
"learning_rate": 0.0,
|
| 249 |
+
"loss": 0.3243,
|
| 250 |
+
"step": 6910
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 10.0,
|
| 254 |
+
"eval_exact_match": 61.9375,
|
| 255 |
+
"eval_f1": 75.6746197198531,
|
| 256 |
+
"eval_runtime": 54.8671,
|
| 257 |
+
"eval_samples_per_second": 74.653,
|
| 258 |
+
"eval_steps_per_second": 2.679,
|
| 259 |
+
"step": 6910
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 10.0,
|
| 263 |
+
"step": 6910,
|
| 264 |
+
"total_flos": 3.788022162461184e+16,
|
| 265 |
+
"train_loss": 0.6408887451878504,
|
| 266 |
+
"train_runtime": 5720.422,
|
| 267 |
+
"train_samples_per_second": 33.789,
|
| 268 |
+
"train_steps_per_second": 1.208
|
| 269 |
+
}
|
| 270 |
+
],
|
| 271 |
+
"logging_steps": 500,
|
| 272 |
+
"max_steps": 6910,
|
| 273 |
+
"num_input_tokens_seen": 0,
|
| 274 |
+
"num_train_epochs": 10,
|
| 275 |
+
"save_steps": 500,
|
| 276 |
+
"total_flos": 3.788022162461184e+16,
|
| 277 |
+
"train_batch_size": 28,
|
| 278 |
+
"trial_name": null,
|
| 279 |
+
"trial_params": null
|
| 280 |
+
}
|