scenario-kd-po-ner-full-mdeberta_data-univner_half44

This model is a fine-tuned version of haryoaw/scenario-TCR-NER_data-univner_half on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 61.3215
  • Precision: 0.7767
  • Recall: 0.7826
  • F1: 0.7796
  • Accuracy: 0.9781

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: 3e-05
  • train_batch_size: 8
  • eval_batch_size: 32
  • seed: 44
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
134.7601 0.5828 500 105.4285 0.6228 0.4135 0.4970 0.9475
96.6208 1.1655 1000 91.3605 0.7044 0.6166 0.6576 0.9679
84.9489 1.7483 1500 84.4145 0.7378 0.7175 0.7275 0.9735
78.4875 2.3310 2000 79.7743 0.7268 0.7713 0.7484 0.9753
73.8037 2.9138 2500 76.5950 0.7425 0.7393 0.7409 0.9752
69.9903 3.4965 3000 73.3846 0.7568 0.7804 0.7684 0.9769
66.64 4.0793 3500 70.9653 0.7632 0.7647 0.7640 0.9772
63.8746 4.6620 4000 68.7757 0.7722 0.7560 0.7640 0.9769
61.8679 5.2448 4500 67.0563 0.7822 0.7667 0.7744 0.9776
60.0989 5.8275 5000 65.7140 0.7687 0.7730 0.7709 0.9772
58.5339 6.4103 5500 64.4640 0.7721 0.7827 0.7774 0.9780
57.5319 6.9930 6000 63.4900 0.7793 0.7768 0.7780 0.9778
56.4947 7.5758 6500 62.8514 0.7706 0.7811 0.7758 0.9776
55.6103 8.1585 7000 62.1384 0.7772 0.7784 0.7778 0.9781
55.1971 8.7413 7500 61.7481 0.7837 0.7784 0.7810 0.9783
54.7227 9.3240 8000 61.4653 0.7765 0.7863 0.7814 0.9784
54.4924 9.9068 8500 61.3215 0.7767 0.7826 0.7796 0.9781

Framework versions

  • Transformers 4.44.2
  • Pytorch 2.1.1+cu121
  • Datasets 2.14.5
  • Tokenizers 0.19.1
Downloads last month
7
Safetensors
Model size
236M params
Tensor type
F32
·
Inference API
Unable to determine this model’s pipeline type. Check the docs .

Model tree for haryoaw/scenario-kd-po-ner-full-mdeberta_data-univner_half44