Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +242 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,242 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 800,
|
| 3 |
+
"best_metric": 0.6842105263157895,
|
| 4 |
+
"best_model_checkpoint": "/home/pricie/trusca/ner/models_ocr/detect_perplex_param/model_lambda0.2_th0.9/checkpoint-800",
|
| 5 |
+
"epoch": 160.0,
|
| 6 |
+
"eval_steps": 200,
|
| 7 |
+
"global_step": 800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 10.0,
|
| 14 |
+
"grad_norm": 0.18833546340465546,
|
| 15 |
+
"learning_rate": 4.902e-05,
|
| 16 |
+
"loss": 0.1822,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 20.0,
|
| 21 |
+
"grad_norm": 0.13318823277950287,
|
| 22 |
+
"learning_rate": 4.8020000000000004e-05,
|
| 23 |
+
"loss": 0.0069,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 30.0,
|
| 28 |
+
"grad_norm": 0.11782731115818024,
|
| 29 |
+
"learning_rate": 4.702e-05,
|
| 30 |
+
"loss": 0.0022,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 40.0,
|
| 35 |
+
"grad_norm": 0.03338160738348961,
|
| 36 |
+
"learning_rate": 4.602e-05,
|
| 37 |
+
"loss": 0.001,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 40.0,
|
| 42 |
+
"eval_LOC": {
|
| 43 |
+
"f1": 0.0,
|
| 44 |
+
"number": 1,
|
| 45 |
+
"precision": 0.0,
|
| 46 |
+
"recall": 0.0
|
| 47 |
+
},
|
| 48 |
+
"eval_PERS": {
|
| 49 |
+
"f1": 0.6796116504854368,
|
| 50 |
+
"number": 56,
|
| 51 |
+
"precision": 0.7446808510638298,
|
| 52 |
+
"recall": 0.625
|
| 53 |
+
},
|
| 54 |
+
"eval_loss": 0.0714135617017746,
|
| 55 |
+
"eval_overall_accuracy": 0.9904264577893821,
|
| 56 |
+
"eval_overall_f1": 0.6603773584905661,
|
| 57 |
+
"eval_overall_precision": 0.7142857142857143,
|
| 58 |
+
"eval_overall_recall": 0.6140350877192983,
|
| 59 |
+
"eval_runtime": 0.1069,
|
| 60 |
+
"eval_samples_per_second": 140.301,
|
| 61 |
+
"eval_steps_per_second": 9.353,
|
| 62 |
+
"step": 200
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 50.0,
|
| 66 |
+
"grad_norm": 0.004488145466893911,
|
| 67 |
+
"learning_rate": 4.502e-05,
|
| 68 |
+
"loss": 0.0006,
|
| 69 |
+
"step": 250
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 60.0,
|
| 73 |
+
"grad_norm": 0.0032296597491949797,
|
| 74 |
+
"learning_rate": 4.402e-05,
|
| 75 |
+
"loss": 0.0005,
|
| 76 |
+
"step": 300
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 70.0,
|
| 80 |
+
"grad_norm": 0.06219864264130592,
|
| 81 |
+
"learning_rate": 4.3020000000000005e-05,
|
| 82 |
+
"loss": 0.0005,
|
| 83 |
+
"step": 350
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 80.0,
|
| 87 |
+
"grad_norm": 0.037636227905750275,
|
| 88 |
+
"learning_rate": 4.202e-05,
|
| 89 |
+
"loss": 0.0004,
|
| 90 |
+
"step": 400
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 80.0,
|
| 94 |
+
"eval_LOC": {
|
| 95 |
+
"f1": 0.0,
|
| 96 |
+
"number": 1,
|
| 97 |
+
"precision": 0.0,
|
| 98 |
+
"recall": 0.0
|
| 99 |
+
},
|
| 100 |
+
"eval_PERS": {
|
| 101 |
+
"f1": 0.7307692307692307,
|
| 102 |
+
"number": 56,
|
| 103 |
+
"precision": 0.7916666666666666,
|
| 104 |
+
"recall": 0.6785714285714286
|
| 105 |
+
},
|
| 106 |
+
"eval_loss": 0.07451974600553513,
|
| 107 |
+
"eval_overall_accuracy": 0.9915868871482448,
|
| 108 |
+
"eval_overall_f1": 0.7238095238095237,
|
| 109 |
+
"eval_overall_precision": 0.7916666666666666,
|
| 110 |
+
"eval_overall_recall": 0.6666666666666666,
|
| 111 |
+
"eval_runtime": 0.0932,
|
| 112 |
+
"eval_samples_per_second": 160.973,
|
| 113 |
+
"eval_steps_per_second": 10.732,
|
| 114 |
+
"step": 400
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 90.0,
|
| 118 |
+
"grad_norm": 0.01358821615576744,
|
| 119 |
+
"learning_rate": 4.1020000000000004e-05,
|
| 120 |
+
"loss": 0.0005,
|
| 121 |
+
"step": 450
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 100.0,
|
| 125 |
+
"grad_norm": 0.0060653723776340485,
|
| 126 |
+
"learning_rate": 4.002e-05,
|
| 127 |
+
"loss": 0.0004,
|
| 128 |
+
"step": 500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 110.0,
|
| 132 |
+
"grad_norm": 0.015429244376718998,
|
| 133 |
+
"learning_rate": 3.902e-05,
|
| 134 |
+
"loss": 0.0003,
|
| 135 |
+
"step": 550
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 120.0,
|
| 139 |
+
"grad_norm": 0.0023402925580739975,
|
| 140 |
+
"learning_rate": 3.802e-05,
|
| 141 |
+
"loss": 0.0003,
|
| 142 |
+
"step": 600
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 120.0,
|
| 146 |
+
"eval_LOC": {
|
| 147 |
+
"f1": 0.0,
|
| 148 |
+
"number": 1,
|
| 149 |
+
"precision": 0.0,
|
| 150 |
+
"recall": 0.0
|
| 151 |
+
},
|
| 152 |
+
"eval_PERS": {
|
| 153 |
+
"f1": 0.6923076923076924,
|
| 154 |
+
"number": 56,
|
| 155 |
+
"precision": 0.75,
|
| 156 |
+
"recall": 0.6428571428571429
|
| 157 |
+
},
|
| 158 |
+
"eval_loss": 0.07376705855131149,
|
| 159 |
+
"eval_overall_accuracy": 0.9907165651290978,
|
| 160 |
+
"eval_overall_f1": 0.6857142857142857,
|
| 161 |
+
"eval_overall_precision": 0.75,
|
| 162 |
+
"eval_overall_recall": 0.631578947368421,
|
| 163 |
+
"eval_runtime": 0.0895,
|
| 164 |
+
"eval_samples_per_second": 167.643,
|
| 165 |
+
"eval_steps_per_second": 11.176,
|
| 166 |
+
"step": 600
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 130.0,
|
| 170 |
+
"grad_norm": 0.09169554710388184,
|
| 171 |
+
"learning_rate": 3.702e-05,
|
| 172 |
+
"loss": 0.0003,
|
| 173 |
+
"step": 650
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 140.0,
|
| 177 |
+
"grad_norm": 0.011466496624052525,
|
| 178 |
+
"learning_rate": 3.6020000000000004e-05,
|
| 179 |
+
"loss": 0.0004,
|
| 180 |
+
"step": 700
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 150.0,
|
| 184 |
+
"grad_norm": 0.0007753117242828012,
|
| 185 |
+
"learning_rate": 3.502e-05,
|
| 186 |
+
"loss": 0.0003,
|
| 187 |
+
"step": 750
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 160.0,
|
| 191 |
+
"grad_norm": 0.0963754802942276,
|
| 192 |
+
"learning_rate": 3.402e-05,
|
| 193 |
+
"loss": 0.0003,
|
| 194 |
+
"step": 800
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 160.0,
|
| 198 |
+
"eval_LOC": {
|
| 199 |
+
"f1": 0.0,
|
| 200 |
+
"number": 1,
|
| 201 |
+
"precision": 0.0,
|
| 202 |
+
"recall": 0.0
|
| 203 |
+
},
|
| 204 |
+
"eval_PERS": {
|
| 205 |
+
"f1": 0.7090909090909091,
|
| 206 |
+
"number": 56,
|
| 207 |
+
"precision": 0.7222222222222222,
|
| 208 |
+
"recall": 0.6964285714285714
|
| 209 |
+
},
|
| 210 |
+
"eval_loss": 0.07110230624675751,
|
| 211 |
+
"eval_overall_accuracy": 0.9912967798085291,
|
| 212 |
+
"eval_overall_f1": 0.7027027027027027,
|
| 213 |
+
"eval_overall_precision": 0.7222222222222222,
|
| 214 |
+
"eval_overall_recall": 0.6842105263157895,
|
| 215 |
+
"eval_runtime": 0.0914,
|
| 216 |
+
"eval_samples_per_second": 164.058,
|
| 217 |
+
"eval_steps_per_second": 10.937,
|
| 218 |
+
"step": 800
|
| 219 |
+
}
|
| 220 |
+
],
|
| 221 |
+
"logging_steps": 50,
|
| 222 |
+
"max_steps": 2500,
|
| 223 |
+
"num_input_tokens_seen": 0,
|
| 224 |
+
"num_train_epochs": 500,
|
| 225 |
+
"save_steps": 200,
|
| 226 |
+
"stateful_callbacks": {
|
| 227 |
+
"TrainerControl": {
|
| 228 |
+
"args": {
|
| 229 |
+
"should_epoch_stop": false,
|
| 230 |
+
"should_evaluate": false,
|
| 231 |
+
"should_log": false,
|
| 232 |
+
"should_save": true,
|
| 233 |
+
"should_training_stop": false
|
| 234 |
+
},
|
| 235 |
+
"attributes": {}
|
| 236 |
+
}
|
| 237 |
+
},
|
| 238 |
+
"total_flos": 1.1915455279104e+16,
|
| 239 |
+
"train_batch_size": 64,
|
| 240 |
+
"trial_name": null,
|
| 241 |
+
"trial_params": null
|
| 242 |
+
}
|