siridech24/wangchanberta-sentiment-fine-tuned

WangchanBERTa Sentiment Classifier

This is a finetuned sentiment classifier based on WangchanBERTa, trained on Thai customer messages from the TISCO Insure platform.

Labels

• Negative
• Neutral
• Positive

Example Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

tokenizer = AutoTokenizer.from_pretrained("siridech24/sentiment")
model = AutoModelForSequenceClassification.from_pretrained("siridech24/sentiment")

inputs = tokenizer("ประทับใจบริการมากครับ", return_tensors="pt")
outputs = model(**inputs)
pred = torch.argmax(outputs.logits, dim=1)
print(pred.item())