File size: 1,457 Bytes
d6ca0c3
 
 
 
33ec535
 
 
aea6cc2
33ec535
 
aea6cc2
 
 
 
 
 
 
 
 
 
 
 
33ec535
 
aea6cc2
33ec535
 
 
 
aea6cc2
33ec535
 
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
---
license: unknown
base_model:
- airesearch/wangchanberta-base-att-spm-uncased
---
# AmbatronBERTa

AmbatronBERTa is a Thai language model fine-tuned specifically for text classification tasks, built upon the WangchanBERTa architecture.

## Model Description
AmbatronBERTa is designed to handle the complexities of the Thai language. It has been fine-tuned on a dataset of over 3,000 research papers to improve classification accuracy. Leveraging the transformer-based WangchanBERTa, it efficiently captures the nuances of Thai text, making it suitable for classifying documents across multiple fields.

## Developers
AmbatronBERTa was developed by students at **King Mongkut's University of Technology North Bangkok**:
- **Peerawat Banpahan**
- **Waris Thongpho**

## Use Cases
AmbatronBERTa can be applied to a wide range of tasks, such as:
- **Research Classification:** Categorizing academic papers into relevant topics.
- **Document Organization:** Classifying articles, blogs, and other documents by themes.
- **Sentiment Analysis:** Analyzing sentiment in Thai-language texts across various contexts.

## How to Use
To use AmbatronBERTa with the `transformers` library:

```python
from transformers import AutoTokenizer, AutoModelForSequenceClassification

# Load the tokenizer and model
tokenizer = AutoTokenizer.from_pretrained("Peerawat2024/AmbatronBERTa")
model = AutoModelForSequenceClassification.from_pretrained("Peerawat2024/AmbatronBERTa")