highlightedreport-classifier-test
This model is a fine-tuned version of latterworks/highlightedreport-classifier-test on the None dataset. It achieves the following results on the evaluation set:
- Accuracy: 0.7840
- Loss: 0.6084
- F1: 0.7701
- Precision: 0.7694
- Recall: 0.7708
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: 1e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Accuracy | Validation Loss | F1 | Precision | Recall |
---|---|---|---|---|---|---|---|
0.6461 | 0.2375 | 100 | 0.7873 | 0.5155 | 0.7772 | 0.7647 | 0.7900 |
0.4988 | 0.4751 | 200 | 0.8014 | 0.4455 | 0.7839 | 0.8012 | 0.7673 |
0.4784 | 0.7126 | 300 | 0.7944 | 0.4508 | 0.7914 | 0.7554 | 0.8310 |
0.4827 | 0.9501 | 400 | 0.7793 | 0.4632 | 0.7836 | 0.7259 | 0.8512 |
0.4628 | 1.1876 | 500 | 0.7944 | 0.4459 | 0.7819 | 0.7787 | 0.7851 |
0.4622 | 1.4252 | 600 | 0.7907 | 0.4591 | 0.7890 | 0.7488 | 0.8338 |
0.4508 | 1.6627 | 700 | 0.7927 | 0.4550 | 0.7875 | 0.7590 | 0.8181 |
0.4572 | 1.9002 | 800 | 0.7903 | 0.4516 | 0.7639 | 0.8104 | 0.7224 |
0.4359 | 2.1378 | 900 | 0.7917 | 0.4772 | 0.7874 | 0.7558 | 0.8217 |
0.3967 | 2.3753 | 1000 | 0.7984 | 0.4567 | 0.7797 | 0.8003 | 0.7601 |
0.415 | 2.6128 | 1100 | 0.7852 | 0.4792 | 0.7847 | 0.7408 | 0.8342 |
0.4097 | 2.8504 | 1200 | 0.7965 | 0.4661 | 0.7847 | 0.7795 | 0.7900 |
0.3962 | 3.0879 | 1300 | 0.7972 | 0.4655 | 0.7738 | 0.8120 | 0.7391 |
0.3738 | 3.3254 | 1400 | 0.7887 | 0.4740 | 0.7806 | 0.7612 | 0.8011 |
0.3618 | 3.5629 | 1500 | 0.7935 | 0.4706 | 0.7720 | 0.8015 | 0.7445 |
0.3604 | 3.8005 | 1600 | 0.7942 | 0.4779 | 0.78 | 0.7828 | 0.7772 |
0.3533 | 4.0380 | 1700 | 0.7892 | 0.4899 | 0.7752 | 0.7760 | 0.7744 |
0.3194 | 4.2755 | 1800 | 0.7902 | 0.5034 | 0.7785 | 0.7717 | 0.7854 |
0.3285 | 4.5131 | 1900 | 0.7893 | 0.4958 | 0.7767 | 0.7730 | 0.7804 |
0.3256 | 4.7506 | 2000 | 0.7908 | 0.4952 | 0.7720 | 0.7905 | 0.7544 |
0.321 | 4.9881 | 2100 | 0.7873 | 0.5050 | 0.7760 | 0.7675 | 0.7847 |
0.2915 | 5.2257 | 2200 | 0.7872 | 0.5167 | 0.7722 | 0.7761 | 0.7683 |
0.2819 | 5.4632 | 2300 | 0.7828 | 0.5344 | 0.7745 | 0.7554 | 0.7947 |
0.2839 | 5.7007 | 2400 | 0.7812 | 0.5529 | 0.7761 | 0.7465 | 0.8082 |
0.2836 | 5.9382 | 2500 | 0.7771 | 0.5433 | 0.7741 | 0.7384 | 0.8135 |
0.2686 | 6.1758 | 2600 | 0.7832 | 0.5545 | 0.7692 | 0.7687 | 0.7698 |
0.2559 | 6.4133 | 2700 | 0.7820 | 0.5578 | 0.7728 | 0.7564 | 0.7900 |
0.2525 | 6.6508 | 2800 | 0.7837 | 0.5647 | 0.7703 | 0.7678 | 0.7730 |
0.251 | 6.8884 | 2900 | 0.7867 | 0.5588 | 0.7713 | 0.7764 | 0.7662 |
0.2463 | 7.1259 | 3000 | 0.7877 | 0.5754 | 0.7738 | 0.7739 | 0.7737 |
0.2284 | 7.3634 | 3100 | 0.7842 | 0.5907 | 0.7758 | 0.7571 | 0.7954 |
0.2295 | 7.6010 | 3200 | 0.7835 | 0.5832 | 0.7654 | 0.7789 | 0.7523 |
0.234 | 7.8385 | 3300 | 0.7807 | 0.5821 | 0.7670 | 0.7650 | 0.7690 |
0.2296 | 8.0760 | 3400 | 0.7850 | 0.5823 | 0.7667 | 0.7813 | 0.7527 |
0.2161 | 8.3135 | 3500 | 0.7837 | 0.5908 | 0.7694 | 0.7699 | 0.7690 |
0.2253 | 8.5511 | 3600 | 0.7857 | 0.5907 | 0.7648 | 0.7887 | 0.7423 |
0.21 | 8.7886 | 3700 | 0.7835 | 0.6021 | 0.7719 | 0.7636 | 0.7804 |
0.2123 | 9.0261 | 3800 | 0.7840 | 0.6025 | 0.7691 | 0.7720 | 0.7662 |
0.1977 | 9.2637 | 3900 | 0.7827 | 0.6081 | 0.7655 | 0.7755 | 0.7559 |
0.2061 | 9.5012 | 4000 | 0.7838 | 0.6090 | 0.7715 | 0.7656 | 0.7776 |
0.2032 | 9.7387 | 4100 | 0.7850 | 0.6081 | 0.7718 | 0.7690 | 0.7747 |
0.2077 | 9.9762 | 4200 | 0.7838 | 0.6084 | 0.7699 | 0.7694 | 0.7705 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.6.0+cu124
- Datasets 3.4.1
- Tokenizers 0.21.1
- Downloads last month
- 38
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support
Model tree for latterworks/highlightedreport-classifier-test
Unable to build the model tree, the base model loops to the model itself. Learn more.