Downloads · 30 days
8
10% of all-time downloads
emotion_detection
aayansh26/emotion_detection
emotion_detection is a machine learning model from aayansh26. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for keras. The card lists the license as mit.
This project is a deep learning-based web application that detects human emotions from images. The model is trained on the FER2013 dataset using transfer learning with ResNet50.
All-time downloads
80
Public
Repo size
285 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.keras285 MB · 100%
At a glance
- Library
- keras
- License
- mit
- Access
- Public
- Created
- Mar 26, 2026
- Updated
- Mar 26, 2026
- SHA
- 8126c96e
Model card
From the Hugging Face model README
😊 Facial Emotion Detection using ResNet50
This project is a deep learning-based web application that detects human emotions from images.
The model is trained on the FER2013 dataset using transfer learning with ResNet50.
📌 Features
-
Detects 7 emotions:
- Angry 😠
- Disgust 🤢
- Fear 😨
- Happy 😄
- Sad 😢
- Surprise 😲
- Neutral 😐
-
Achieved ~68% test accuracy
-
Uses ResNet50 (Transfer Learning)
-
Real-time image upload and prediction
-
Deployed using Flask + Render
-
Model hosted on Hugging Face
🧠 Model Details
- Base Model: ResNet50 (pretrained on ImageNet)
- Input Shape:
(224, 224, 3) - Preprocessing: Pixel normalization (
/255) - Loss Function: Categorical Crossentropy
- Optimizer: Adam
- Techniques Used:
- Data Augmentation
- Class Weighting
- EarlyStopping
- ReduceLROnPlateau
- ModelCheckpoint
📊 Performance
- Train Accuracy: ~75%
- Test Accuracy: ~68%
🖼️ Pipeline
- Library
- keras
- License
- mit
- Created
- Mar 26, 2026
- Updated
- Mar 26, 2026