| --- |
| language: en |
| license: apache-2.0 |
| tags: |
| - text-classification |
| - emotion |
| - emotion-detection |
| - affective-computing |
| - distilbert |
| - transformers |
| - pytorch |
| pipeline_tag: text-classification |
| widget: |
| - text: "I was skeptical at first, but the results completely blew me away with how good it turned out!" |
| example_title: "Joy Example" |
| - text: "I have been sitting here in complete darkness, feeling utterly exhausted and defeated." |
| example_title: "Sadness Example" |
| - text: "Get out of my face right now before I lose my temper completely!" |
| example_title: "Anger Example" |
| - text: "Every single moment spent with you makes my life feel infinitely more meaningful." |
| example_title: "Love Example" |
| --- |
| |
| # DistilBERT Multiclass Emotion Classifier 🎭 |
|
|
| A fine-tuned **DistilBERT (Distilled Bidirectional Encoder Representations from Transformers)** sequence classification model trained to detect nuanced affective and emotional states from unstructured natural language. |
|
|
| - **Author**: [Jayan Gupta](https://huggingface.co/Jayan12321) |
|
|
| --- |
|
|
| ## 🎯 Model Overview & Classes |
| The model classifies text across 6 primary emotional dimensions: |
| * `0`: **Sadness** (grief, sorrow, exhaustion) |
| * `1`: **Joy** (happiness, triumph, gratitude) |
| * `2`: **Love** (affection, admiration, intimacy) |
| * `3`: **Anger** (irritation, frustration, rage) |
| * `4`: **Fear** (anxiety, dread, panic) |
| * `5`: **Surprise** (shock, amazement, disbelief) |
|
|
| --- |
|
|
| ## 💻 Python Quickstart with Transformers Pipeline |
|
|
| ```python |
| from transformers import pipeline |
| |
| # Load fine-tuned pipeline directly from Hugging Face |
| classifier = pipeline( |
| "text-classification", |
| model="Jayan12321/distilbert-emotion-classifier" |
| ) |
| |
| # Run inference |
| sample_text = "I received the scholarship offer this morning and I cannot stop smiling!" |
| prediction = classifier(sample_text) |
| |
| print(prediction) |
| # Output: [{'label': 'joy', 'score': 0.987}] |
| ``` |
|
|
