khanhnamle1994

def calculate_loss(self, x, y):
    assert len(x) == len(y)
    output = Softmax()
    layers = self.forward_propagation(x)
    loss = 0.0
    for i, layer in enumerate(layers):
        loss += output.loss(layer.mulv, y[i])
    return loss / float(len(y))

def calculate_total_loss(self, X, Y):

khanhnamle1994

def bptt(self, x, y):
    assert len(x) == len(y)
    output = Softmax()
    layers = self.forward_propagation(x)
    dU = np.zeros(self.U.shape)
    dV = np.zeros(self.V.shape)
    dW = np.zeros(self.W.shape)

    T = len(layers)
    prev_s_t = np.zeros(self.hidden_dim)

khanhnamle1994

def sgd(self, x, y, learning_rate):
    dU, dW, dV = self.bptt(x, y)
    self.U -= learning_rate * dU
    self.V -= learning_rate * dV
    self.W -= learning_rate * dW

khanhnamle1994

TaylorSwift_vc <- paste(as.character(PlaylistSongs$id[1:38]), sep="", collapse=",")
ArianaGrande_vc <- paste(as.character(PlaylistSongs$id[39:86]), sep="", collapse=",")
KendrickLamar_vc <- paste(as.character(PlaylistSongs$id[87:124]), sep="", collapse=",")
ShawnMendes_vc <- paste(as.character(PlaylistSongs$id[125:177]), sep="", collapse=",")
Maroon5_vc <- paste(as.character(PlaylistSongs$id[178:226]), sep="", collapse=",")
PostMalone_vc <- paste(as.character(PlaylistSongs$id[227:261]), sep="", collapse=",")
Kygo_vc <- paste(as.character(PlaylistSongs$id[262:299]), sep="", collapse=",")
TheChainsmokers_vc <- paste(as.character(PlaylistSongs$id[300:333]), sep="", collapse=",")
Adele_vc <- paste(as.character(PlaylistSongs$id[334:358]), sep="", collapse=",")
Drake_vc <- paste(as.character(PlaylistSongs$id[359:408]), sep="", collapse=",")

khanhnamle1994

# Obtain the audio features for each musician
TaylorSwift <- get_features(TaylorSwift_vc)
ArianaGrande <- get_features(ArianaGrande_vc)
KendrickLamar <- get_features(KendrickLamar_vc)
ShawnMendes <- get_features(ShawnMendes_vc)
Maroon5 <- get_features(Maroon5_vc)
PostMalone <- get_features(PostMalone_vc)
Kygo <- get_features(Kygo_vc)
TheChainsmokers <- get_features(TheChainsmokers_vc)
Adele <- get_features(Adele_vc)

khanhnamle1994

library(radarchart)
library(tidyr)
sample1 <- mean_features[mean_features$Musician %in% c("TaylorSwift", "ArianaGrande", "KendrickLamar", "ShawnMendes", "Maroon5", "PostMalone", "Kygo", "TheChainsmokers", "Adele", "Drake"),]

mean_features_norm_1 <- cbind(sample1[1], apply(sample1[-1],2, function(x){(x-min(x)) / diff(range(x))})) 

radarDF_1 <- gather(mean_features_norm_1, key=Attribute, value=Score, -Musician) %>%
  spread(key=Musician, value=Score)

chartJSRadar(scores = radarDF_1, scaleStartValue = -1, maxScale =1, showToolTipLabel = TRUE)

khanhnamle1994

# K-Means Clustering
library(ggfortify)
library(ggthemes)
set.seed(5000)

k_means <- kmeans(scaled.features, 6)

kmeans_plot <- autoplot(k_means, 
                        main = "K-means Clustering", 
                        data = scaled.features,

khanhnamle1994

# Cluster Radar Chart
mean_features_norm_50 <- cbind(mean_features[1], apply(mean_features[-1],2,scale))

cluster_centers <- as.data.frame(k_means$centers)
cluster <- c("Cluster 1", "Cluster 2", "Cluster 3", "Cluster 4", "Cluster 5", "Cluster 6")
cluster_centers <- cbind(cluster, cluster_centers)

radarDF_6 <- gather(cluster_centers, key=Attribute, value=Score, -cluster) %>%
  spread(key=cluster, value=Score)
# Change the colours according to clusters

khanhnamle1994

# Danceability feature
library(ggplot2)
danceability_subset <- mean_features_norm_50[,c("Musician","danceability", "cluster")]
danceability_subset <- danceability_subset[order(danceability_subset$danceability, decreasing = TRUE), ]

danceability_plot <- ggplot(danceability_subset, 
                            aes(x = reorder(Musician, danceability), 
                                y = danceability, label = danceability)) + 
  xlab("Musician") + ylab("Danceability") + 
  geom_bar(stat = 'identity', aes(fill = cluster), width = .5) + 

khanhnamle1994

# Import libraries
from keras.utils import to_categorical
from sklearn.model_selection import train_test_split

# Load training and test data into dataframes
data_train = pd.read_csv('data/fashion-mnist_train.csv')
data_test = pd.read_csv('data/fashion-mnist_test.csv')

# X forms the training images, and y forms the training labels
X = np.array(data_train.iloc[:, 1:])