-
This repository contains every relevant file that was used to achieve the objective of training an ML model with audio from Audioset Dataset.
-
This repository does not include the files required to download the dataset; for that, please visit audioset-processing.
To run populate_dataset.py you need to have the input directory as the following image:
Each subfolder contains the preprocessed image in .png format that was obtained with thesis_preprocessing.ipynb.