A Comparative Analysis of Lab View and PyTorch for Machine Learning: The gap between Experimentation and Production

Abstract- This paper presents a comparative analysis of handwritten digit recognition performance between LabVIEW and PyTorch frameworks, utilizing a Convolutional Neural Network (CNN). The model is designed to classify digits from the MNIST dataset, which consists of 28×28 grayscale images of handwritten digits (0–9). The dataset includes 60,000 training images and 10,000 test images, providing a standardized benchmark for evaluating model performance. Metrics such as accuracy, training time, memory usage, and inference speed are evaluated. The results provide insight into the strengths and weaknesses of these frameworks in terms of efficiency, scalability, and usability. Results indicate that while both frameworks are effective, PyTorch offers faster training and inference, whereas LabVIEW demonstrates marginally better training accuracy.