Qwen2.5-1.5B Comment Classifier

Fine-tuned Qwen/Qwen2.5-1.5B for 4-class comment sentiment classification.

Classes

Label Description
positive Admiration, joy, love, gratitude, etc.
negative Anger, disgust, sadness, fear, etc.
neutral Neutral statements
ambiguous Surprise, confusion, curiosity, realization

Training Details

  • Base model: Qwen/Qwen2.5-1.5B
  • Method: LoRA (r=16, alpha=32)
  • Dataset: GoEmotions (58k+ comments)
  • Epochs: 3

Usage

from transformers import pipeline

classifier = pipeline("text-classification", model="jovincia/qwen25-comment-classifier")
result = classifier("This product is amazing!")
print(result)

Performance Metrics

See the results/ folder for detailed evaluation and monitoring reports.

Downloads last month
14
Safetensors
Model size
2B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Dataset used to train jovincia/qwen25-comment-classifier

Space using jovincia/qwen25-comment-classifier 1