In [12]:
import pandas as pd
import numpy as np
import os
from sklearn.preprocessing import MinMaxScaler,StandardScaler,normalize
import matplotlib.pyplot as plt 
#importing the header files 
from collections import Counter

from sklearn.manifold import TSNE
import time
import sklearn
from sklearn.decomposition import PCA 

from sklearn import mixture


from sklearn.preprocessing import MinMaxScaler,StandardScaler,LabelEncoder
from sklearn.model_selection import GridSearchCV

import plotly.express as px
from collections import Counter 
from plotly.subplots import make_subplots
import plotly.graph_objects as go
import umap
from audio_results_util import load_dataset,plotClusters,encodedLabels,plotData3d,plotData2d,plotAllClusterModels,actualDistribution
%matplotlib inline
In [2]:
train_data,true_labels,true_encoded_labels=load_dataset('Audio_features_vowel/mfcc_muslim_features_vowel.csv')
In [3]:
train_data.shape
Out[3]:
(240, 13)
In [4]:
results_df=pd.read_csv('audio-results4/mfcc_muslim_features_vowel.csv')
results_df
Out[4]:
Unnamed: 0 n_clusters ARI AMI H C V FM A R P silhouette calinski davies
0 K-Means 5 0.043502 9.051057e-02 0.148992 0.095706 0.116548 0.347453 0.195833 0.143850 0.192945 0.111764 25.306967 2.165053
1 Agglomerative clustering 2 -0.082096 2.692399e-02 0.028813 0.058807 0.038676 0.535644 0.491667 0.148428 0.122917 0.091499 23.309823 2.359739
2 Birch 2 0.065935 6.978901e-02 0.048229 0.308017 0.083399 0.694086 0.650000 0.196226 0.135065 0.175546 8.455691 1.938029
3 DBSCAN 4 -0.035317 5.199972e-03 0.038441 0.039130 0.038782 0.455758 0.175000 0.044025 0.116667 -0.094465 4.567208 3.598599
4 Mean-shift 1 0.000000 -5.797594e-17 0.000000 1.000000 0.000000 0.694318 0.662500 0.200000 0.132500 -1.000000 -1.000000 -1.000000
5 Optics 1 0.000000 -5.797594e-17 0.000000 1.000000 0.000000 0.694318 0.662500 0.200000 0.132500 -1.000000 -1.000000 -1.000000
6 Gaussian-mixture 5 -0.012643 7.748358e-02 0.124180 0.092258 0.105865 0.367038 0.045833 0.045087 0.158319 0.104357 20.759759 2.128010
In [5]:
actualDistribution(train_data,true_labels,true_encoded_labels)
plotAllClusterModels(train_data,'audio-results4/mfcc_muslim_features_vowel.csv')

Actual Distribution of the labels in the dataset: --> 
Counter({'a': 159, 'u': 47, 'cantdecide': 13, 'misaligned': 13, 'other': 8})
Counter({0: 159, 4: 47, 1: 13, 2: 13, 3: 8})

audio-results4/mfcc_muslim_features_vowel.csv_kmeans_labels.npy
K-Means
predicted_labels--> Counter({3: 58, 4: 52, 2: 51, 0: 48, 1: 31})
 2D representation

 3D representation

Agglomerative
predicted_labels--> Counter({0: 192, 1: 48})
 2D representation

 3D representation

Birch
predicted_labels--> Counter({0: 231, 1: 9})
 2D representation

 3D representation

DBSCAN
predicted_labels--> Counter({-1: 152, 0: 60, 2: 13, 1: 11, 3: 4})
 2D representation

 3D representation

OPTICS
predicted_labels--> Counter({0: 240})
 2D representation

 3D representation

MEAN-SHIFT
predicted_labels--> Counter({0: 240})
 2D representation

 3D representation

Gaussian-Mixture
predicted_labels--> Counter({2: 106, 3: 56, 1: 45, 4: 22, 0: 11})
 2D representation

 3D representation
In [ ]:

PCA transformed data

In [ ]:
In [6]:
results_df=pd.read_csv('audio-results4/mfcc_muslim_features_vowel.csv-pca.csv')
results_df
Out[6]:
Unnamed: 0 n_clusters ARI AMI H C V FM A R P silhouette calinski davies
0 K-Means 5 0.069847 1.123442e-01 0.176939 0.112579 0.137606 0.364503 0.283333 0.186113 0.251920 0.291529 104.014995 1.039882
1 Agglomerative clustering 4 0.036628 8.460775e-02 0.123520 0.092454 0.105753 0.378131 0.162500 0.138921 0.105588 0.268633 92.939300 1.069304
2 Birch 4 0.035732 9.620200e-02 0.137252 0.101868 0.116942 0.373868 0.183333 0.069473 0.165901 0.295274 104.835271 1.021278
3 DBSCAN 2 -0.031591 -7.034795e-04 0.016707 0.023991 0.019697 0.508949 0.458333 0.115304 0.107212 0.092514 10.799876 3.914580
4 Mean-shift 1 0.000000 -5.797594e-17 0.000000 1.000000 0.000000 0.694318 0.662500 0.200000 0.132500 -1.000000 -1.000000 -1.000000
5 Optics 1 0.034321 4.918684e-04 0.009043 0.049637 0.015298 0.676448 0.020833 0.005241 0.075758 -0.070527 5.408569 1.704391
6 Gaussian-mixture 5 0.072831 1.073271e-01 0.170433 0.108763 0.132787 0.368108 0.158333 0.186008 0.166298 0.285470 102.003188 1.059069
In [7]:
pca_transformed=PCA(n_components=3).fit_transform(train_data)
pca_transformed=pd.DataFrame(pca_transformed)
actualDistribution(pca_transformed,true_labels,true_encoded_labels)
plotAllClusterModels(pca_transformed,'audio-results4/mfcc_muslim_features_vowel.csv-pca')

Actual Distribution of the labels in the dataset: --> 
Counter({'a': 159, 'u': 47, 'cantdecide': 13, 'misaligned': 13, 'other': 8})
Counter({0: 159, 4: 47, 1: 13, 2: 13, 3: 8})

audio-results4/mfcc_muslim_features_vowel.csv-pca_kmeans_labels.npy
K-Means
predicted_labels--> Counter({0: 56, 4: 49, 2: 47, 3: 45, 1: 43})
 2D representation

 3D representation

Agglomerative
predicted_labels--> Counter({0: 79, 1: 61, 2: 56, 3: 44})
 2D representation

 3D representation

Birch
predicted_labels--> Counter({2: 75, 3: 57, 1: 55, 0: 53})
 2D representation

 3D representation

DBSCAN
predicted_labels--> Counter({0: 171, -1: 60, 1: 9})
 2D representation

 3D representation

OPTICS
predicted_labels--> Counter({-1: 229, 0: 11})
 2D representation

 3D representation

MEAN-SHIFT
predicted_labels--> Counter({0: 240})
 2D representation

 3D representation

Gaussian-Mixture
predicted_labels--> Counter({2: 59, 3: 51, 4: 45, 0: 45, 1: 40})
 2D representation

 3D representation
In [ ]:

t-SNE transformed data

In [8]:
results_df=pd.read_csv('audio-results4/mfcc_muslim_features_vowel.csv-tsne.csv')
results_df
Out[8]:
Unnamed: 0 n_clusters ARI AMI H C V FM A R P silhouette calinski davies
0 K-Means 5 0.029578 7.315166e-02 0.128054 0.081358 0.099500 0.331949 0.266667 0.313387 0.268207 0.239667 84.677258 1.118155
1 Agglomerative clustering 240 0.000000 -6.363593e-14 1.000000 0.186350 0.314157 0.000000 0.008333 0.000347 0.008333 -1.000000 -1.000000 -1.000000
2 Birch 240 0.000000 -6.363593e-14 1.000000 0.186350 0.314157 0.000000 0.008333 0.000409 0.008333 -1.000000 -1.000000 -1.000000
3 DBSCAN 2 0.038171 3.479782e-03 0.018938 0.026523 0.022098 0.522857 0.450000 0.113208 0.116129 -0.008944 4.914734 5.616258
4 Mean-shift 2 0.038426 3.452892e-02 0.027243 0.211054 0.048257 0.688987 0.650000 0.196226 0.133906 0.170785 12.665392 1.093261
5 Optics 1 0.000000 -5.797594e-17 0.000000 1.000000 0.000000 0.694318 0.662500 0.200000 0.132500 -1.000000 -1.000000 -1.000000
6 Gaussian-mixture 5 0.041849 6.003447e-02 0.110762 0.071837 0.087151 0.350904 0.300000 0.138525 0.238275 0.224458 79.367738 1.109322
In [9]:
tsne_transformed=TSNE(n_components=3, n_jobs=-1).fit_transform(train_data)
tsne_transformed=pd.DataFrame(tsne_transformed)
actualDistribution(tsne_transformed,true_labels,true_encoded_labels)

plotAllClusterModels(train_data,'audio-results4/mfcc_muslim_features_vowel.csv-tsne')

Actual Distribution of the labels in the dataset: --> 
Counter({'a': 159, 'u': 47, 'cantdecide': 13, 'misaligned': 13, 'other': 8})
Counter({0: 159, 4: 47, 1: 13, 2: 13, 3: 8})

audio-results4/mfcc_muslim_features_vowel.csv-tsne_kmeans_labels.npy
K-Means
predicted_labels--> Counter({2: 52, 3: 49, 4: 49, 0: 47, 1: 43})
 2D representation

 3D representation

Agglomerative
Too many labels to show
Birch
Too many labels to show
DBSCAN
predicted_labels--> Counter({0: 155, -1: 80, 1: 5})
 2D representation

 3D representation

OPTICS
predicted_labels--> Counter({0: 240})
 2D representation

 3D representation

MEAN-SHIFT
predicted_labels--> Counter({0: 233, 1: 7})
 2D representation

 3D representation

Gaussian-Mixture
predicted_labels--> Counter({0: 67, 2: 58, 4: 45, 1: 36, 3: 34})
 2D representation

 3D representation
In [ ]:

Umap transformed data

In [10]:
results_df=pd.read_csv('audio-results4/mfcc_muslim_features_vowel.csv-umap.csv')
results_df
Out[10]:
Unnamed: 0 n_clusters ARI AMI H C V FM A R P silhouette calinski davies
0 K-Means 5 0.047288 0.086824 0.144898 0.092420 0.112857 0.347715 0.195833 0.227957 0.189806 0.398901 165.997345 0.857070
1 Agglomerative clustering 5 0.056063 0.101537 0.163350 0.104050 0.127124 0.354022 0.166667 0.213425 0.175005 0.394911 161.757610 0.902572
2 Birch 11 0.033347 0.093476 0.239972 0.103299 0.144428 0.248680 0.137500 0.046010 0.117084 0.374132 162.586606 0.917346
3 DBSCAN 16 0.002862 0.053740 0.221093 0.088475 0.126377 0.219612 0.070833 0.010444 0.059040 0.165878 48.466406 1.069505
4 Mean-shift 3 0.010153 0.079324 0.097528 0.092807 0.095109 0.415867 0.241667 0.185970 0.118136 0.300556 104.579573 1.152263
5 Optics 3 -0.094513 0.053072 0.079895 0.078367 0.079124 0.415691 0.108333 0.062571 0.165785 0.149905 51.665732 1.157557
6 Gaussian-mixture 5 0.056762 0.090600 0.149421 0.095575 0.116581 0.356505 0.220833 0.283642 0.225959 0.396812 164.830310 0.835609
In [11]:
umap_transformed= umap.UMAP(random_state=42,n_components=3).fit_transform(train_data)
umap_transformed=pd.DataFrame(umap_transformed)
actualDistribution(umap_transformed,true_labels,true_encoded_labels)
plotAllClusterModels(umap_transformed,'audio-results4/mfcc_muslim_features_vowel.csv-umap')

Actual Distribution of the labels in the dataset: --> 
Counter({'a': 159, 'u': 47, 'cantdecide': 13, 'misaligned': 13, 'other': 8})
Counter({0: 159, 4: 47, 1: 13, 2: 13, 3: 8})

audio-results4/mfcc_muslim_features_vowel.csv-umap_kmeans_labels.npy
K-Means
predicted_labels--> Counter({1: 57, 0: 52, 2: 47, 4: 45, 3: 39})
 2D representation

 3D representation

Agglomerative
predicted_labels--> Counter({3: 55, 4: 50, 1: 49, 0: 47, 2: 39})
 2D representation

 3D representation

Birch
predicted_labels--> Counter({1: 31, 4: 27, 0: 25, 2: 24, 3: 23, 7: 23, 5: 20, 9: 19, 6: 19, 8: 16, 10: 13})
 2D representation

 3D representation

DBSCAN
predicted_labels--> Counter({-1: 52, 6: 33, 11: 20, 2: 19, 0: 17, 1: 16, 5: 12, 13: 12, 3: 10, 8: 9, 10: 8, 4: 7, 12: 6, 7: 6, 9: 5, 15: 5, 14: 3})
 2D representation

 3D representation

OPTICS
predicted_labels--> Counter({-1: 153, 1: 39, 0: 27, 2: 21})
 2D representation

 3D representation

MEAN-SHIFT
predicted_labels--> Counter({0: 106, 2: 68, 1: 66})
 2D representation

 3D representation

Gaussian-Mixture
predicted_labels--> Counter({4: 62, 1: 47, 3: 47, 0: 46, 2: 38})
 2D representation

 3D representation
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]:
In [ ]: