metadata

license: mit
tags:
  - generated_from_trainer
datasets:
  - null
metrics:
  - accuracy
model-index:
  - name: BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-finetuned-pubmedqa-1
    results:
      - task:
          name: Text Classification
          type: text-classification
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.52

BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext-finetuned-pubmedqa-1

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: 1.0977
Accuracy: 0.52

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: 1e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	57	0.9515	0.54
No log	2.0	114	0.9530	0.54
No log	3.0	171	0.9705	0.46
No log	4.0	228	0.9688	0.56
No log	5.0	285	0.9674	0.58
No log	6.0	342	0.9474	0.58
No log	7.0	399	1.0309	0.54
No log	8.0	456	1.0307	0.56
0.6657	9.0	513	1.0825	0.52
0.6657	10.0	570	1.0977	0.52

Framework versions

Transformers 4.10.2
Pytorch 1.9.0+cu102
Datasets 1.12.0
Tokenizers 0.10.3