The advent of low-cost wearable Internet of Things (IoT) sensors have made it possible to continuously acquire physiological signals such as the electrocardiogram (ECG) for long durations. Techniques for automated analysis is essential for deriving intelligence from such a large quantity of data. This paper presents a 1-dimensional convolutional neural network (CNN) for heartbeat classification from ECG signal obtained from an ambulatory device. The proposed technique can classify heartbeats into 5 classes as specified in the AAMI standard and was tested using the Physionet MIT-BIH Arrhythmia database. To address the imbalance of classes in the dataset we used SMOTE algorithm to augment the dataset. The network was trained using the augmented data and achieved an accuracy of 98.12%, sensitivity of 98.07% and a specificity of 98.29%.