Skip to content

Instantly share code, notes, and snippets.

View jaron's full-sized avatar

Jaron Collis jaron

33 followers · 5 following
View GitHub Profile
@jaron
jaron / urban-sound-cnn-1.py
Created October 4, 2016 15:10
A stacked convolutional neural network (CNN) to classify the Urban Sound 8K dataset. Uses Tensorflow, with Keras to provide some higher-level abstractions. Inspired by a blog post by Aaqib Saeed (http://aqibsaeed.github.io/2016-09-24-urban-sound-classification-part-2/)
import numpy as np
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation, Flatten
from keras.layers import Convolution2D, MaxPooling2D
from keras.optimizers import Adam
from keras.utils import np_utils
from sklearn import metrics
# to run this code, you'll need to load the following data:
# train_x, train_y
@jaron
jaron / urban-sound-cnn-salamon.py
Last active July 12, 2019 15:07
A Keras/Tensorflow implementation of the 5-layer CNN described in Salamon and Bello's paper (https://arxiv.org/pdf/1608.04363.pdf). See http://aqibsaeed.github.io/2016-09-24-urban-sound-classification-part-2/ for a description on how to create the data this uses.
import numpy as np
from keras.models import Sequential
from keras.layers import Dense, Dropout, Activation, Flatten
from keras.layers import Convolution2D, MaxPooling2D
from keras.optimizers import SGD
from keras.regularizers import l2, activity_l2
from keras.utils import np_utils
from sklearn import metrics
# to run this code, you'll need to load the following data: