THIS CODE IS DEPRECATED AND HIGHLY SUGGEST USE THE LATEST CODE HERE:

https://github.com/qiuqiangkong/audioset_classification

===========================================================

Audio Set classification with attention model: A probabilistic perspective

Author: Qiuqiang Kong (q.kong@surrey.ac.uk) and Yong Xu (yong.xu@surrey.ac.uk) contribute equally to this work, Wenwu Wang (w.wang@surrey.ac.uk) and Mark D. Plumbley (m.plumbley@surrey.ac.uk)

This is the souce code for Audio Set classification, an implementation of this paper https://arxiv.org/abs/1711.00927

Download dataset

We convert the tensorflow type data to numpy data and stored in hdf5 file. The hdf5 data can be downloaded here https://drive.google.com/open?id=0B49XSFgf-0yVQk01eG92RHg4WTA

Install requirements

pip install -r requirements.txt

Run

./runme.sh