File size: 2,436 Bytes
7ed27e3 9e883c8 7ed27e3 9e883c8 7ed27e3 9e883c8 7ed27e3 9e883c8 7ed27e3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 |
---
base_model: medicalai/ClinicalBERT
tags:
- generated_from_trainer
metrics:
- f1
- accuracy
model-index:
- name: working
results: []
---
<!-- 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](https://huggingface.co/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
| Training Loss | Epoch | Step | Validation Loss | F1 | Roc Auc | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:|:--------:|
| No log | 1.0 | 136 | 0.0223 | 0.9834 | 0.9930 | 0.9853 |
| No log | 2.0 | 272 | 0.0163 | 0.9881 | 0.9959 | 0.9926 |
| No log | 3.0 | 408 | 0.0137 | 0.9834 | 0.9930 | 0.9853 |
| 0.0216 | 4.0 | 544 | 0.0120 | 0.9834 | 0.9930 | 0.9853 |
| 0.0216 | 5.0 | 680 | 0.0108 | 0.9834 | 0.9930 | 0.9853 |
### Framework versions
- Transformers 4.37.0
- Pytorch 2.1.2+cu118
- Datasets 2.16.1
- Tokenizers 0.15.0
|