stake52/emotion-distilbert

Emotion Classification with DistilBERT

Model Description

A DistilBERT model fine-tuned for six-class English emotion classification.

Base Model

distilbert/distilbert-base-uncased

Dataset

dair-ai/emotion

The dataset contains six emotion classes: sadness, joy, love, anger, fear, and surprise.

Task

Text classification.

Usage

from transformers import pipeline

classifier = pipeline(
    "text-classification",
    model="stake52/emotion-distilbert"
)

result = classifier("I am very happy today!")
print(result)

Evaluation

The model was evaluated using accuracy and macro-F1 on the held-out test split. Report the exact values from the completed run.

Limitations

Emotion classification is subjective and text can contain multiple or ambiguous emotions. Predictions should not be interpreted as reliable measurements of a person's mental state.

Intended Use

Educational projects, NLP experimentation, and research.