Upload trainer_state.json
Browse files- trainer_state.json +163 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,163 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.16890692710876465,
|
| 3 |
+
"best_model_checkpoint": "./indic_ur/checkpoint-6250",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 6250,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.4,
|
| 12 |
+
"learning_rate": 1.885714285714286e-05,
|
| 13 |
+
"loss": 0.8064,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.8,
|
| 18 |
+
"learning_rate": 1.7714285714285717e-05,
|
| 19 |
+
"loss": 0.5114,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 1.0,
|
| 24 |
+
"eval_LOC_f1": 0.8276643990929706,
|
| 25 |
+
"eval_ORG_f1": 0.6985534297713485,
|
| 26 |
+
"eval_PER_f1": 0.6433615819209039,
|
| 27 |
+
"eval_loss": 0.41389304399490356,
|
| 28 |
+
"eval_overall_accuracy": 0.8477015558698727,
|
| 29 |
+
"eval_overall_f1": 0.7343858431644692,
|
| 30 |
+
"eval_overall_precision": 0.7192863211554801,
|
| 31 |
+
"eval_overall_recall": 0.7501329080276449,
|
| 32 |
+
"eval_runtime": 7.4875,
|
| 33 |
+
"eval_samples_per_second": 133.557,
|
| 34 |
+
"eval_steps_per_second": 8.414,
|
| 35 |
+
"step": 1250
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.2,
|
| 39 |
+
"learning_rate": 1.6571428571428574e-05,
|
| 40 |
+
"loss": 0.4147,
|
| 41 |
+
"step": 1500
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.6,
|
| 45 |
+
"learning_rate": 1.542857142857143e-05,
|
| 46 |
+
"loss": 0.3518,
|
| 47 |
+
"step": 2000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 2.0,
|
| 51 |
+
"learning_rate": 1.4285714285714287e-05,
|
| 52 |
+
"loss": 0.3117,
|
| 53 |
+
"step": 2500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 2.0,
|
| 57 |
+
"eval_LOC_f1": 0.875703564727955,
|
| 58 |
+
"eval_ORG_f1": 0.8162930563149262,
|
| 59 |
+
"eval_PER_f1": 0.8648328896776102,
|
| 60 |
+
"eval_loss": 0.25353190302848816,
|
| 61 |
+
"eval_overall_accuracy": 0.9142857142857143,
|
| 62 |
+
"eval_overall_f1": 0.8532248075732106,
|
| 63 |
+
"eval_overall_precision": 0.8519434628975265,
|
| 64 |
+
"eval_overall_recall": 0.8545100124047492,
|
| 65 |
+
"eval_runtime": 7.6113,
|
| 66 |
+
"eval_samples_per_second": 131.384,
|
| 67 |
+
"eval_steps_per_second": 8.277,
|
| 68 |
+
"step": 2500
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.4,
|
| 72 |
+
"learning_rate": 1.3142857142857145e-05,
|
| 73 |
+
"loss": 0.2388,
|
| 74 |
+
"step": 3000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.8,
|
| 78 |
+
"learning_rate": 1.2e-05,
|
| 79 |
+
"loss": 0.2468,
|
| 80 |
+
"step": 3500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 3.0,
|
| 84 |
+
"eval_LOC_f1": 0.8871085214857976,
|
| 85 |
+
"eval_ORG_f1": 0.8523093447905478,
|
| 86 |
+
"eval_PER_f1": 0.8841001747233547,
|
| 87 |
+
"eval_loss": 0.21500876545906067,
|
| 88 |
+
"eval_overall_accuracy": 0.9272277227722773,
|
| 89 |
+
"eval_overall_f1": 0.8747007182761373,
|
| 90 |
+
"eval_overall_precision": 0.8753993610223643,
|
| 91 |
+
"eval_overall_recall": 0.8740031897926634,
|
| 92 |
+
"eval_runtime": 7.4281,
|
| 93 |
+
"eval_samples_per_second": 134.624,
|
| 94 |
+
"eval_steps_per_second": 8.481,
|
| 95 |
+
"step": 3750
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 3.2,
|
| 99 |
+
"learning_rate": 1.0857142857142858e-05,
|
| 100 |
+
"loss": 0.2071,
|
| 101 |
+
"step": 4000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 3.6,
|
| 105 |
+
"learning_rate": 9.714285714285715e-06,
|
| 106 |
+
"loss": 0.1688,
|
| 107 |
+
"step": 4500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 4.0,
|
| 111 |
+
"learning_rate": 8.571428571428571e-06,
|
| 112 |
+
"loss": 0.1752,
|
| 113 |
+
"step": 5000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 4.0,
|
| 117 |
+
"eval_LOC_f1": 0.902119552274351,
|
| 118 |
+
"eval_ORG_f1": 0.870843000773395,
|
| 119 |
+
"eval_PER_f1": 0.8982737361282367,
|
| 120 |
+
"eval_loss": 0.20536212623119354,
|
| 121 |
+
"eval_overall_accuracy": 0.936987270155587,
|
| 122 |
+
"eval_overall_f1": 0.890302066772655,
|
| 123 |
+
"eval_overall_precision": 0.8874801901743264,
|
| 124 |
+
"eval_overall_recall": 0.8931419457735247,
|
| 125 |
+
"eval_runtime": 7.3438,
|
| 126 |
+
"eval_samples_per_second": 136.17,
|
| 127 |
+
"eval_steps_per_second": 8.579,
|
| 128 |
+
"step": 5000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 4.4,
|
| 132 |
+
"learning_rate": 7.428571428571429e-06,
|
| 133 |
+
"loss": 0.1226,
|
| 134 |
+
"step": 5500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 4.8,
|
| 138 |
+
"learning_rate": 6.285714285714286e-06,
|
| 139 |
+
"loss": 0.1241,
|
| 140 |
+
"step": 6000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 5.0,
|
| 144 |
+
"eval_LOC_f1": 0.9250525332710716,
|
| 145 |
+
"eval_ORG_f1": 0.9045801526717557,
|
| 146 |
+
"eval_PER_f1": 0.9203379601689801,
|
| 147 |
+
"eval_loss": 0.16890692710876465,
|
| 148 |
+
"eval_overall_accuracy": 0.9524045261669024,
|
| 149 |
+
"eval_overall_f1": 0.9169995561473592,
|
| 150 |
+
"eval_overall_precision": 0.9187122020633227,
|
| 151 |
+
"eval_overall_recall": 0.9152932837143364,
|
| 152 |
+
"eval_runtime": 7.4827,
|
| 153 |
+
"eval_samples_per_second": 133.642,
|
| 154 |
+
"eval_steps_per_second": 8.419,
|
| 155 |
+
"step": 6250
|
| 156 |
+
}
|
| 157 |
+
],
|
| 158 |
+
"max_steps": 8750,
|
| 159 |
+
"num_train_epochs": 7,
|
| 160 |
+
"total_flos": 212723461717536.0,
|
| 161 |
+
"trial_name": null,
|
| 162 |
+
"trial_params": null
|
| 163 |
+
}
|