NLP_HerbalMultilabelClassification

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

Herbal Multilabel Classification

This model is a fine-tuned version of medicalai/ClinicalBERT on a custom dataset. It achieves the following results on the evaluation set:

• Loss: 0.0108
• F1: 0.9834
• Roc Auc: 0.9930
• Accuracy: 0.9853

Model description

It is a multilabel classification model that deals with 10 herbal plants (Jackfruit, Sambong, Lemon, Jasmine, Mango, Mint, Ampalaya, Malunggay, Guava, Lagundi) which are abundant in the Philippines. The model classifies a herbal(s) that is/are applicable based on the input symptom of the user.

Intended uses & limitations

The model is created for the purpose of completing a University course. It will be integrated to a React Native mobile application for the project. The model performs well when the input of the user contains a symptom that has been trained to the model from the dataset. However, other words/inputs that do not present a significance to the purpose of the model would generate an underwhelming and inaccurate result.

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-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: 5

Training results

Framework versions

• Transformers 4.37.0
• Pytorch 2.1.2+cu118
• Datasets 2.16.1
• Tokenizers 0.15.0