scenario-kd-scr-ner-full-mdeberta-halfen_data-univner_en66

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: 392.9234
  • Precision: 0.5127
  • Recall: 0.3975
  • F1: 0.4478
  • Accuracy: 0.9594

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: 66
  • 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
603.5902 1.28 500 532.0310 0.7568 0.0290 0.0558 0.9414
497.4866 2.55 1000 479.9597 0.2891 0.1646 0.2098 0.9463
451.2688 3.83 1500 443.6332 0.4884 0.2619 0.3410 0.9526
420.4378 5.1 2000 421.6972 0.4431 0.3789 0.4085 0.9567
400.9749 6.38 2500 408.6700 0.5306 0.3230 0.4015 0.9574
388.4506 7.65 3000 398.0555 0.5026 0.4037 0.4478 0.9596
380.4155 8.93 3500 392.9234 0.5127 0.3975 0.4478 0.9594

Framework versions

  • Transformers 4.33.3
  • Pytorch 2.1.1+cu121
  • Datasets 2.14.5
  • Tokenizers 0.13.3
Downloads last month
11
Inference API
Unable to determine this model's library. Check the docs .

Model tree for haryoaw/scenario-kd-scr-ner-full-mdeberta-halfen_data-univner_en66