This project focuses on emotion detection using a Convolutional Neural Network (CNN). The CNN model is trained to detect seven different emotions from grayscale images with dimensions 227x227 pixels. The emotions include happiness, sadness, anger, surprise, fear, disgust, and neutral.
- Python 3.x
- PyTorch
- matplotlib
- NumPy
-
Clone this repository:
git clone https://github.com/your_username/Emotion_detection.git -
Navigate to the project directory:
cd Emotion_detection -
Ensure that all required libraries are installed(if not already):
pip install -r requirements.txt -
Prepare your dataset:
- Organize your grayscale images into folders corresponding to each emotion.
- Ensure that each image has dimensions of 227x227 pixels.
-
Training and testing the model:
- ipynb file contains code for both training and testing the model.
- the model can be loaded as well. My trained model is provided as well.