Hugging Face's logo

Brizape
/
multiCorp_2e-05_LabelNorm_0404

Token Classification
Transformers
PyTorch
bert
Generated from Trainer
Inference Endpoints
Model card Files Community
1
multiCorp_2e-05_LabelNorm_0404 / README.md
Brizape's picture
Brizape
update model card README.md
a0d09e3
preview code
|
raw history blame contribute delete
No virus
4.64 kB
metadata
license: mit
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: multiCorp_2e-05_LabelNorm_0404
    results: []

multiCorp_2e-05_LabelNorm_0404

This model is a fine-tuned version of microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0307
  • Precision: 0.7335
  • Recall: 0.5525
  • F1: 0.6303
  • Accuracy: 0.9910

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
  • training_steps: 2000

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.4089 0.08 25 0.1001 0.0 0.0 0.0 0.9838
0.0776 0.15 50 0.0771 0.0 0.0 0.0 0.9838
0.0576 0.23 75 0.0631 0.5929 0.0521 0.0959 0.9843
0.0621 0.31 100 0.0582 0.2212 0.0179 0.0331 0.9839
0.0438 0.39 125 0.0505 0.4326 0.3346 0.3774 0.9859
0.047 0.46 150 0.0479 0.5205 0.3549 0.4220 0.9868
0.043 0.54 175 0.0461 0.5706 0.3144 0.4054 0.9871
0.0292 0.62 200 0.0437 0.4402 0.3899 0.4135 0.9865
0.0395 0.7 225 0.0411 0.5338 0.4669 0.4981 0.9882
0.0345 0.77 250 0.0414 0.5533 0.3471 0.4266 0.9869
0.0491 0.85 275 0.0379 0.6573 0.3447 0.4523 0.9883
0.0388 0.93 300 0.0370 0.6529 0.3704 0.4727 0.9884
0.0348 1.01 325 0.0371 0.5327 0.5191 0.5258 0.9883
0.0316 1.08 350 0.0363 0.5613 0.4988 0.5282 0.9884
0.0252 1.16 375 0.0340 0.6533 0.4957 0.5637 0.9898
0.0386 1.24 400 0.0367 0.5861 0.5829 0.5845 0.9889
0.0251 1.32 425 0.0362 0.6452 0.4444 0.5263 0.9890
0.0337 1.39 450 0.0348 0.6794 0.4981 0.5748 0.9896
0.0306 1.47 475 0.0371 0.7112 0.4350 0.5398 0.9895
0.022 1.55 500 0.0340 0.7126 0.5556 0.6244 0.9907
0.0292 1.63 525 0.0306 0.6797 0.5533 0.6100 0.9907
0.0277 1.7 550 0.0321 0.6529 0.5782 0.6133 0.9903
0.0295 1.78 575 0.0313 0.6564 0.5992 0.6265 0.9906
0.0253 1.86 600 0.0351 0.7402 0.4545 0.5632 0.9902
0.0228 1.93 625 0.0304 0.668 0.6498 0.6588 0.9910
0.0276 2.01 650 0.0313 0.6880 0.5183 0.5912 0.9904
0.0185 2.09 675 0.0325 0.6661 0.6257 0.6453 0.9907
0.0199 2.17 700 0.0303 0.6809 0.6459 0.6629 0.9911
0.0191 2.24 725 0.0307 0.6933 0.6156 0.6521 0.9910
0.0167 2.32 750 0.0334 0.6620 0.5930 0.6256 0.9906
0.0247 2.4 775 0.0317 0.6591 0.6062 0.6315 0.9902
0.0236 2.48 800 0.0315 0.7354 0.5798 0.6484 0.9914
0.0191 2.55 825 0.0367 0.7523 0.4420 0.5569 0.9900
0.0252 2.63 850 0.0307 0.7335 0.5525 0.6303 0.9910

Framework versions

  • Transformers 4.27.4
  • Pytorch 2.0.0+cu118
  • Datasets 2.11.0
  • Tokenizers 0.13.3