Edit model card

tmvar_0.0001_ES2

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: 0.0187
  • Precision: 0.8449
  • Recall: 0.8541
  • F1: 0.8495
  • Accuracy: 0.9961

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: 0.0001
  • 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: 1000

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
0.2263 1.47 25 0.0788 0.0 0.0 0.0 0.9843
0.0492 2.94 50 0.0355 0.2576 0.3676 0.3029 0.9863
0.0258 4.41 75 0.0224 0.6 0.6811 0.6380 0.9933
0.013 5.88 100 0.0141 0.8267 0.9027 0.8630 0.9969
0.0031 7.35 125 0.0162 0.8218 0.8973 0.8579 0.9971
0.0028 8.82 150 0.0187 0.8449 0.8541 0.8495 0.9961

Framework versions

  • Transformers 4.27.4
  • Pytorch 2.0.0+cu118
  • Datasets 2.11.0
  • Tokenizers 0.13.2
Downloads last month
7