Dataset Preprocessing for Internal Combustion Engine Fault Classification (Dataset Preprocessing)

This code presents the implementation of a pre-processing pipeline for the database of an internal combustion engine of a car that will be used for fault detection using a Deep Learning model. The process begins with setting up Google Drive to access the dataset, importing the necessary libraries for audio processing and machine learning, and defining a function to extract MFCC coefficients from fixed-padding audio files. Next, it prepares the data, extracting the features from the audio files located in the specified path, organizes them into NumPy arrays, normalizes the features, encodes the class labels using LabelEncoder, divides the data into training, validation, and test sets using stratification to ensure that the distribution of classes is similar in each set.