|
|
--- |
|
|
license: mit |
|
|
datasets: |
|
|
- google-research-datasets/go_emotions |
|
|
language: |
|
|
- en |
|
|
base_model: |
|
|
- microsoft/MiniLM-L12-H384-uncased |
|
|
tags: |
|
|
- emotion |
|
|
- emotion classification |
|
|
- multilabel-classification |
|
|
- goemotions |
|
|
- mental-health |
|
|
- journaling |
|
|
- transformers |
|
|
- minilm |
|
|
--- |
|
|
|
|
|
# 🧠 MindMap Emotion Classifier v3 - MiniLM Version (microsoft/MiniLM-L12-H384-uncased + GoEmotions) |
|
|
|
|
|
## Note: This model is part of the experiment to find the best-performing emotion classification model for our digital journaling web application called MindMap. |
|
|
|
|
|
A fine-tuned multi-label emotion classification model based on [microsoft/MiniLM-L12-H384-uncased](https://huggingface.co/microsoft/MiniLM-L12-H384-uncased) and trained on the [GoEmotions](https://github.com/google-research/goemotions) dataset. This model is designed to power emotional tagging for personal journaling and mental wellness applications like **MindMap**. |
|
|
|
|
|
## 🚀 Model Details |
|
|
|
|
|
- **Base Model**: MiniLM-L12-H384-uncased |
|
|
- **Task**: Multi-label emotion classification |
|
|
- **Dataset**: [GoEmotions](https://huggingface.co/datasets/go_emotions) (27 emotions + neutral) |
|
|
- **Output**: Probability scores for each of the 28 emotion labels |
|
|
|
|
|
### 🏷️ Supported Emotions (28 classes): |
|
|
admiration, amusement, anger, annoyance, approval, caring, confusion, curiosity, desire, |
|
|
disappointment, disapproval, disgust, embarrassment, excitement, fear, gratitude, grief, |
|
|
joy, love, nervousness, optimism, pride, realization, relief, remorse, sadness, surprise |
|
|
|
|
|
|
|
|
## 📊 Evaluation |
|
|
This model was evaluated using: |
|
|
|
|
|
Metrics: F1-score (micro/macro), Precision, Recall |
|
|
|
|
|
Validation split: 90/10 on the simplified GoEmotions dataset |
|
|
|
|
|
Threshold: 0.05 for emotion label activation |
|
|
|
|
|
## 🧠 Use Case |
|
|
Originally used for MindMap, a digital journaling app that helps users track and reflect on their emotional well-being. The model enables emotion-aware feedback and visualizations, offering therapeutic insight to users based on their writing. |
|
|
|
|
|
## 📦 Model Files |
|
|
model.safetensors: Model weights |
|
|
|
|
|
config.json: Model configuration |
|
|
|
|
|
tokenizer.json, tokenizer_config.json: Tokenizer details |
|
|
|
|
|
special_tokens_map.json, vocab.json: Tokenizer vocabulary |
|
|
|
|
|
|
|
|
## 📚 Citation / Credit |
|
|
Base model: Microsoft MiniLM-L12-H384-uncased |
|
|
|
|
|
Dataset: GoEmotions by Google Research |
|
|
|
|
|
🛠 Maintained by @wncelrcn |
