--- license: mit tags: - generated_from_trainer metrics: - precision - recall - f1 - accuracy model-index: - name: gpt2-ner-invoiceSenderRecipient_all_inv_03_01 results: [] --- # gpt2-ner-invoiceSenderRecipient_all_inv_03_01 This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the None dataset. It achieves the following results on the evaluation set: - Loss: 0.0418 - Precision: 0.7250 - Recall: 0.7797 - F1: 0.7514 - Accuracy: 0.9858 ## Model description More information needed ## Intended uses & limitations More information needed ## Training and evaluation data More information needed ## Training procedure ### Training hyperparameters The following hyperparameters were used during training: - learning_rate: 2e-05 - train_batch_size: 4 - eval_batch_size: 4 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 1 - mixed_precision_training: Native AMP ### Training results | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |:-------------:|:-----:|:-----:|:---------------:|:---------:|:------:|:------:|:--------:| | 0.5038 | 0.01 | 500 | 0.1148 | 0.3694 | 0.1907 | 0.2515 | 0.9648 | | 0.1126 | 0.02 | 1000 | 0.0921 | 0.4770 | 0.3930 | 0.4309 | 0.9709 | | 0.0965 | 0.04 | 1500 | 0.0834 | 0.5167 | 0.4808 | 0.4981 | 0.9733 | | 0.0885 | 0.05 | 2000 | 0.0844 | 0.5038 | 0.5091 | 0.5064 | 0.9719 | | 0.0834 | 0.06 | 2500 | 0.0745 | 0.5463 | 0.5648 | 0.5554 | 0.9753 | | 0.0796 | 0.07 | 3000 | 0.0725 | 0.5587 | 0.6160 | 0.5860 | 0.9762 | | 0.0756 | 0.09 | 3500 | 0.0675 | 0.5892 | 0.6110 | 0.5999 | 0.9778 | | 0.0779 | 0.1 | 4000 | 0.0677 | 0.5844 | 0.6275 | 0.6052 | 0.9776 | | 0.0709 | 0.11 | 4500 | 0.0655 | 0.5932 | 0.6365 | 0.6141 | 0.9782 | | 0.0703 | 0.12 | 5000 | 0.0630 | 0.6090 | 0.6296 | 0.6191 | 0.9789 | | 0.0695 | 0.14 | 5500 | 0.0623 | 0.6160 | 0.6417 | 0.6286 | 0.9791 | | 0.0657 | 0.15 | 6000 | 0.0608 | 0.6268 | 0.6437 | 0.6351 | 0.9798 | | 0.0662 | 0.16 | 6500 | 0.0620 | 0.6046 | 0.6916 | 0.6452 | 0.9792 | | 0.0647 | 0.17 | 7000 | 0.0578 | 0.6504 | 0.6454 | 0.6479 | 0.9809 | | 0.0642 | 0.19 | 7500 | 0.0580 | 0.6484 | 0.6564 | 0.6524 | 0.9807 | | 0.0644 | 0.2 | 8000 | 0.0579 | 0.6379 | 0.6606 | 0.6491 | 0.9804 | | 0.0605 | 0.21 | 8500 | 0.0564 | 0.6336 | 0.7112 | 0.6702 | 0.9810 | | 0.0621 | 0.22 | 9000 | 0.0556 | 0.6503 | 0.6982 | 0.6734 | 0.9813 | | 0.0608 | 0.24 | 9500 | 0.0549 | 0.6679 | 0.6692 | 0.6686 | 0.9816 | | 0.0593 | 0.25 | 10000 | 0.0547 | 0.6560 | 0.7122 | 0.6830 | 0.9818 | | 0.06 | 0.26 | 10500 | 0.0550 | 0.6258 | 0.7449 | 0.6802 | 0.9811 | | 0.0574 | 0.27 | 11000 | 0.0539 | 0.6496 | 0.7138 | 0.6802 | 0.9816 | | 0.0574 | 0.28 | 11500 | 0.0540 | 0.6595 | 0.6971 | 0.6778 | 0.9822 | | 0.0578 | 0.3 | 12000 | 0.0527 | 0.6615 | 0.7173 | 0.6883 | 0.9821 | | 0.0569 | 0.31 | 12500 | 0.0518 | 0.6690 | 0.7175 | 0.6924 | 0.9825 | | 0.0556 | 0.32 | 13000 | 0.0512 | 0.6869 | 0.6927 | 0.6898 | 0.9830 | | 0.0539 | 0.33 | 13500 | 0.0513 | 0.6771 | 0.7090 | 0.6927 | 0.9828 | | 0.0547 | 0.35 | 14000 | 0.0505 | 0.6732 | 0.7272 | 0.6992 | 0.9829 | | 0.0564 | 0.36 | 14500 | 0.0493 | 0.6950 | 0.6920 | 0.6935 | 0.9835 | | 0.0549 | 0.37 | 15000 | 0.0505 | 0.6587 | 0.7543 | 0.7033 | 0.9828 | | 0.0552 | 0.38 | 15500 | 0.0495 | 0.6735 | 0.7415 | 0.7059 | 0.9832 | | 0.0511 | 0.4 | 16000 | 0.0498 | 0.6814 | 0.7377 | 0.7085 | 0.9835 | | 0.055 | 0.41 | 16500 | 0.0479 | 0.6896 | 0.7359 | 0.7120 | 0.9838 | | 0.0513 | 0.42 | 17000 | 0.0483 | 0.7121 | 0.7034 | 0.7077 | 0.9839 | | 0.0534 | 0.43 | 17500 | 0.0479 | 0.6925 | 0.7387 | 0.7148 | 0.9839 | | 0.0515 | 0.45 | 18000 | 0.0475 | 0.7003 | 0.7243 | 0.7121 | 0.9840 | | 0.0535 | 0.46 | 18500 | 0.0490 | 0.6747 | 0.7627 | 0.7160 | 0.9833 | | 0.0512 | 0.47 | 19000 | 0.0502 | 0.6667 | 0.7738 | 0.7163 | 0.9830 | | 0.051 | 0.48 | 19500 | 0.0465 | 0.7113 | 0.7251 | 0.7181 | 0.9844 | | 0.0508 | 0.5 | 20000 | 0.0468 | 0.6893 | 0.7652 | 0.7253 | 0.9841 | | 0.0497 | 0.51 | 20500 | 0.0462 | 0.7069 | 0.7469 | 0.7264 | 0.9844 | | 0.0491 | 0.52 | 21000 | 0.0462 | 0.6969 | 0.7608 | 0.7274 | 0.9843 | | 0.0497 | 0.53 | 21500 | 0.0465 | 0.6972 | 0.7569 | 0.7258 | 0.9843 | | 0.0515 | 0.55 | 22000 | 0.0463 | 0.7035 | 0.7538 | 0.7278 | 0.9845 | | 0.0505 | 0.56 | 22500 | 0.0461 | 0.6983 | 0.7625 | 0.7290 | 0.9844 | | 0.0514 | 0.57 | 23000 | 0.0450 | 0.7183 | 0.7391 | 0.7285 | 0.9848 | | 0.0489 | 0.58 | 23500 | 0.0445 | 0.7174 | 0.7520 | 0.7343 | 0.9849 | | 0.0499 | 0.59 | 24000 | 0.0451 | 0.7085 | 0.7577 | 0.7323 | 0.9847 | | 0.052 | 0.61 | 24500 | 0.0458 | 0.6978 | 0.7701 | 0.7321 | 0.9843 | | 0.0471 | 0.62 | 25000 | 0.0452 | 0.7085 | 0.7642 | 0.7353 | 0.9847 | | 0.0478 | 0.63 | 25500 | 0.0449 | 0.7176 | 0.7566 | 0.7365 | 0.9849 | | 0.0472 | 0.64 | 26000 | 0.0443 | 0.7301 | 0.7331 | 0.7316 | 0.9851 | | 0.0479 | 0.66 | 26500 | 0.0444 | 0.7119 | 0.7625 | 0.7363 | 0.9849 | | 0.048 | 0.67 | 27000 | 0.0460 | 0.6895 | 0.7891 | 0.7359 | 0.9843 | | 0.0484 | 0.68 | 27500 | 0.0443 | 0.7145 | 0.7608 | 0.7369 | 0.9849 | | 0.0489 | 0.69 | 28000 | 0.0437 | 0.7122 | 0.7716 | 0.7407 | 0.9851 | | 0.0461 | 0.71 | 28500 | 0.0435 | 0.7140 | 0.7702 | 0.7410 | 0.9853 | | 0.0486 | 0.72 | 29000 | 0.0429 | 0.7230 | 0.7635 | 0.7427 | 0.9854 | | 0.0487 | 0.73 | 29500 | 0.0434 | 0.7225 | 0.7594 | 0.7405 | 0.9853 | | 0.0473 | 0.74 | 30000 | 0.0429 | 0.7219 | 0.7686 | 0.7446 | 0.9855 | | 0.0462 | 0.76 | 30500 | 0.0429 | 0.7175 | 0.7772 | 0.7461 | 0.9854 | | 0.0484 | 0.77 | 31000 | 0.0427 | 0.7174 | 0.7724 | 0.7439 | 0.9855 | | 0.0486 | 0.78 | 31500 | 0.0436 | 0.7076 | 0.7851 | 0.7443 | 0.9850 | | 0.0455 | 0.79 | 32000 | 0.0425 | 0.7187 | 0.7775 | 0.7469 | 0.9856 | | 0.0463 | 0.81 | 32500 | 0.0427 | 0.7160 | 0.7826 | 0.7478 | 0.9855 | | 0.0479 | 0.82 | 33000 | 0.0430 | 0.7141 | 0.7842 | 0.7475 | 0.9853 | | 0.046 | 0.83 | 33500 | 0.0423 | 0.7243 | 0.7690 | 0.7460 | 0.9856 | | 0.0464 | 0.84 | 34000 | 0.0420 | 0.7289 | 0.7659 | 0.7469 | 0.9858 | | 0.0463 | 0.85 | 34500 | 0.0423 | 0.7194 | 0.7813 | 0.7490 | 0.9856 | | 0.0459 | 0.87 | 35000 | 0.0427 | 0.7149 | 0.7872 | 0.7493 | 0.9855 | | 0.0435 | 0.88 | 35500 | 0.0420 | 0.7219 | 0.7759 | 0.7479 | 0.9857 | | 0.0473 | 0.89 | 36000 | 0.0419 | 0.7216 | 0.7812 | 0.7502 | 0.9857 | | 0.0456 | 0.9 | 36500 | 0.0423 | 0.7151 | 0.7892 | 0.7503 | 0.9856 | | 0.0441 | 0.92 | 37000 | 0.0426 | 0.7147 | 0.7899 | 0.7504 | 0.9855 | | 0.0461 | 0.93 | 37500 | 0.0416 | 0.7272 | 0.7754 | 0.7505 | 0.9859 | | 0.0441 | 0.94 | 38000 | 0.0417 | 0.7243 | 0.7793 | 0.7508 | 0.9858 | | 0.0442 | 0.95 | 38500 | 0.0416 | 0.7237 | 0.7812 | 0.7514 | 0.9858 | | 0.0452 | 0.97 | 39000 | 0.0418 | 0.7250 | 0.7797 | 0.7514 | 0.9858 | | 0.0422 | 0.98 | 39500 | 0.0420 | 0.7225 | 0.7835 | 0.7518 | 0.9857 | | 0.0467 | 0.99 | 40000 | 0.0419 | 0.7222 | 0.7839 | 0.7518 | 0.9857 | ### Framework versions - Transformers 4.22.0 - Pytorch 1.10.0 - Datasets 2.3.2 - Tokenizers 0.12.1