bert-base-uncased-Research_Articles_Multilabel

This model is a fine-tuned version of bert-base-uncased.

It achieves the following results on the evaluation set:

• Loss: 0.2039
• F1: 0.8405
• Roc Auc: 0.8976
• Accuracy: 0.7082

Model description

Here is the link to my code for this model: https://github.com/DunnBC22/NLP_Projects/blob/main/Multilabel%20Classification/Research%20Articles/Research%20Articles%20-%20Multilabel%20Classification%20-%20Bert-Base-Uncased.ipynb

Intended uses & limitations

This model could be used to read labels with printed text. You are more than welcome to use it, but remember that it is at your own risk/peril.

Training and evaluation data

Dataset Source: https://www.kaggle.com/datasets/shivanandmn/multilabel-classification-dataset

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
0.2425	1.0	2097	0.1948	0.8348	0.8921	0.7067
0.1739	2.0	4194	0.1986	0.8348	0.8926	0.7072
0.1328	3.0	6291	0.2039	0.8405	0.8976	0.7082

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3