enoriega's picture
update model card README.md
421264a
metadata
license: mit
tags:
  - generated_from_trainer
datasets:
  - enoriega/keyword_pubmed
metrics:
  - accuracy
model-index:
  - name: kw_pubmed_vanilla_sentence_10000_0.0003_2
    results:
      - task:
          name: Masked Language Modeling
          type: fill-mask
        dataset:
          name: enoriega/keyword_pubmed sentence
          type: enoriega/keyword_pubmed
          args: sentence
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.6767448105720579

kw_pubmed_vanilla_sentence_10000_0.0003_2

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

  • Loss: 1.5883
  • Accuracy: 0.6767

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: 0.0003
  • train_batch_size: 16
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 500
  • total_train_batch_size: 8000
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3.0

Training results

Framework versions

  • Transformers 4.18.0
  • Pytorch 1.11.0
  • Datasets 2.1.0
  • Tokenizers 0.12.1