This project focuses on handwritten digit recognition using the MNIST dataset and a Neural Network. The goal is to develop a system that can accurately detect and classify scanned images of handwritten digits.
The MNIST dataset is used, which consists of 60,000 training images and 10,000 testing images. Each image is a 28x28 grayscale image representing a handwritten digit from 0 to 9.
The project is implemented using Python and the Keras library.
-
Data Preprocessing: The input images are reshaped into a 1D array and normalized to have pixel values in the range of 0-1. The target labels are one-hot encoded.
-
Neural Network Model: I created a Sequential model with multiple layers, including Dense layers and Dropout layers to prevent overfitting.
-
Model Training: The model is compiled with an appropriate loss function and optimizer. It is then trained on the training dataset with a specified number of epochs and batch size.
-
Model Evaluation: The trained model is evaluated on the testing dataset to measure its accuracy in recognizing handwritten digits.
-
Visualization: The accuracy and loss curves are plotted to visualize the model's performance. Additionally, a subset of input images with their corresponding labels is displayed to provide a visual representation of the dataset.
To run this project locally, follow these steps:
-
Clone the repository:
git clone https://github.com/your-username/number-recognition.git -
Install the necessary dependencies:
pip install -r requirements.txt -
Run the jupyter notebook.
The model achieves an accuracy of 98.12% on the testing dataset, showcasing its effectiveness in recognizing handwritten digits.
This project was completed by me alone as part of the internship program offered by Bharat Intern. I would like to thank Bharat Intern for providing me with the opportunity to work on this assignment.