PhoBERT-Emotion: Emotion Recognition for Vietnamese Text

This model is a fine-tuned version of vinai/phobert-base on the VSMEC dataset for emotion recognition in Vietnamese text. It achieves competitive performance on this task.

Model Details

• Base Model: vinai/phobert-base
• Dataset: VSMEC (Vietnamese Social Media Emotion Corpus)
• Fine-tuning Framework: HuggingFace Transformers
• Hyperparameters:
  • Batch size: 32
  • Learning rate: 5e-5
  • Epochs: 100
  • Max sequence length: 256

Dataset

The model was trained on the VSMEC dataset, which contains Vietnamese social media text annotated with emotion labels. The dataset includes the following emotion categories: {"Anger": 0, "Disgust": 1, "Enjoyment": 2, "Fear": 3, "Other": 4, "Sadness": 5, "Surprise": 6}.

Results

The model was evaluated using the following metrics:

• Accuracy: <INSERT_ACCURACY>
• F1 Score: <INSERT_F1_SCORE>

Usage

You can use this model for emotion recognition in Vietnamese text. Below is an example of how to use it with the HuggingFace Transformers library:

from transformers import AutoTokenizer, AutoModelForSequenceClassification

tokenizer = AutoTokenizer.from_pretrained("visolex/phobert-emotion")
model = AutoModelForSequenceClassification.from_pretrained("visolex/phobert-emotion")

text = "Tôi rất vui vì hôm nay trời đẹp!"
inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=256)
outputs = model(**inputs)
predicted_class = outputs.logits.argmax(dim=-1).item()

print(f"Predicted emotion: {predicted_class}")