README.md

metadata

language:
  - en
tags:
  - generated_from_trainer
datasets:
  - ade_corpus_v2
metrics:
  - accuracy
model-index:
  - name: >-
      BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-finetuned-ade-v2-classification
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: ade_corpus_v2
          type: ade_corpus_v2
          config: null
          split: None
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9610969387755102

BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-finetuned-ade-v2-classification

This model is a fine-tuned version of BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext on the ade_corpus_v2 dataset. It achieves the following results on the evaluation set:

• Loss: 0.1982
• Accuracy: 0.9611

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: 1.8069489920382708e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 10
• 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	Accuracy
0.1657	1.0	1176	0.1405	0.9511
0.1019	2.0	2352	0.1767	0.9575
0.055	3.0	3528	0.1982	0.9611
0.0424	4.0	4704	0.2038	0.9605

Framework versions

• Transformers 4.25.1
• Pytorch 1.12.1+cu113
• Datasets 2.7.1
• Tokenizers 0.13.2