Flamenco43's picture
End of training
2991715 verified
metadata
license: mit
base_model: microsoft/deberta-v3-base
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: DeBERTaV3800abstractsNER
    results: []

DeBERTaV3800abstractsNER

This model is a fine-tuned version of microsoft/deberta-v3-base on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1966
  • Precision: 0.8035
  • Recall: 0.8626
  • F1: 0.8320
  • Accuracy: 0.9414
  • Per Tag Metrics: {'B-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9932602444284687}, 'I-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9964503953989935}, 'B-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.996989575844716}, 'B-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9949676491732566}, 'B-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9974388928828181}, 'I-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.999370956146657}, 'I-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9952372393961179}, 'B-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9963605319913731}, 'I-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9972591660675773}, 'I-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9961808051761323}, 'B-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9919572250179727}, 'I-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.994832854061826}, 'I-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9849928109273903}, 'B-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9852624011502517}, 'O': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9621675053918045}}

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: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy Per Tag Metrics
No log 1.0 221 0.2739 0.7075 0.7756 0.7400 0.9147 {'B-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9854421279654925}, 'I-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.995057512580877}, 'B-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9960460100647016}, 'B-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.993754493170381}, 'B-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9954169662113588}, 'I-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.999370956146657}, 'I-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.993305176132279}, 'B-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9942487419122933}, 'I-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9955517613227893}, 'I-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.994383537023724}, 'B-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9875988497483824}, 'I-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9914180445722501}, 'I-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9809938892882818}, 'B-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9803199137311287}, 'O': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9465312724658519}}
No log 2.0 442 0.2200 0.7776 0.8219 0.7992 0.9299 {'B-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9920021567217829}, 'I-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9962257368799425}, 'B-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9968098490294752}, 'B-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.994608195542775}, 'B-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9969446441409058}, 'I-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.999370956146657}, 'I-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9944734004313444}, 'B-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9958662832494608}, 'I-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9965851905104242}, 'I-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9958213515456507}, 'B-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.988812005751258}, 'I-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9881829618979152}, 'I-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9851725377426312}, 'B-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9845434938892883}, 'O': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9543943206326384}}
0.3866 3.0 663 0.2019 0.7969 0.8559 0.8253 0.9390 {'B-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9923166786484543}, 'I-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9962706685837527}, 'B-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.996764917325665}, 'B-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9946531272465852}, 'B-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9972142343637671}, 'I-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.999370956146657}, 'I-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9950125808770669}, 'B-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9962257368799425}, 'I-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9970794392523364}, 'I-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9961808051761323}, 'B-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9919572250179727}, 'I-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9940690150970525}, 'I-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9850826743350107}, 'B-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9854421279654925}, 'O': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9603253055355859}}
0.3866 4.0 884 0.1966 0.8035 0.8626 0.8320 0.9414 {'B-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9932602444284687}, 'I-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9964503953989935}, 'B-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.996989575844716}, 'B-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9949676491732566}, 'B-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9974388928828181}, 'I-SPL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.999370956146657}, 'I-CMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9952372393961179}, 'B-APL': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9963605319913731}, 'I-DSC': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9972591660675773}, 'I-SMT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9961808051761323}, 'B-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9919572250179727}, 'I-MAT': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.994832854061826}, 'I-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9849928109273903}, 'B-PRO': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9852624011502517}, 'O': {'precision': 0.0, 'recall': 0.0, 'f1': 0.0, 'accuracy': 0.9621675053918045}}

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.2.1+cu118
  • Datasets 2.19.1
  • Tokenizers 0.19.1