Handwritten Digit Recognition Using CNN in Python with Keras-Tuner

In recent years, deep learning-particularly Convolutional Neural Networks (CNNs)-has emerged as a fundamental tool in pattern recognition and computer vision. One of the most notable applications of CNNs is handwritten digit recognition, which plays a significant role in domains such as banking, education, and document processing. This study employs a CNN architecture to classify handwritten digits from the widely used MNIST dataset. To optimize model performance, the Keras-Tuner library is utilized for automated hyperparameter tuning, including parameters such as the number of layers, number of neurons, learning rate, and activation functions. The results indicate that intelligent selection of hyperparameters significantly enhances model accuracy, and that Keras-Tuner greatly simplifies the iterative experimentation process. The optimized model achieved a test accuracy of ۹۹.۵۸%, demonstrating the effectiveness of the proposed approach in handwritten digit classification tasks.

کلیدواژه ها:

Convolutional Neural Network (CNN) ، Handwritten Digit Recognition ، MNIST Dataset ، Hyperparameter Tuning ، Keras-Tuner