![\"\"](\"https:\/\/i0.wp.com\/bangla.sitestree.com\/wp-content\/uploads\/2024\/05\/image-40.png\" "\\"image-40-png\\"")
<\/a><\/p>\n#!\/usr\/bin\/env python<\/p>\n
coding: utf-8<\/h1>\nIn[1]:<\/h1>\nk-means clustering<\/h1>\n
k-means clustering<\/h1>\n
from numpy import unique
\nfrom numpy import where
\nfrom sklearn.datasets import make_classification
\nfrom sklearn.cluster import KMeans
\nfrom matplotlib import pyplot
\nimport numpy as np
\nimport pandas as pd
\nimport seaborn as sns
\nimport matplotlib.pyplot as plt
\nget_ipython().run_line_magic(‘matplotlib’, ‘inline’)
\nimport pandas as pd
\nimport numpy as np
\nimport numpy as np
\nfrom sklearn.cluster import KMeans
\nfrom sklearn import datasets
\nfrom sklearn.preprocessing import StandardScaler<\/p>\n
In[2]:<\/h1>\n
import warnings
\nwarnings.filterwarnings(‘ignore’)<\/p>\n
In[3]:<\/h1>\nthe combined data<\/h1>\n
data_folder = ‘.\/nhanes_input_data\/’<\/p>\n
import the CSV as a pandas dataframe<\/h1>\n
df = pd.read_csv( data_folder + ‘0_dietaryIntakeDataForClassificationAndAnalysisData.csv’)
\ndf.shape<\/p>\n
In[4]:<\/h1>\n
df.head(5)<\/p>\n
In[5]:<\/h1>\nparameters to be used for KMeans clustring: centres<\/h1>\nX and\/or kdf will have only features we want to create cluster around<\/h1>\n
X and\/or kdf will have only features we want to create cluster around<\/h1>\n
kdf = df[
\u00a0 \u00a0
\u00a0 \u00a0 [
\u00a0 \u00a0 \u00a0 \u00a0 'RIDAGEYR_Age_in_years_at_screening'
\u00a0 \u00a0 \u00a0 \u00a0 ,'URDACT_Albumin_creatinine_ratio_mg_g'
\u00a0 \u00a0 ]
\n]
\nX = kdf
\nX[:5]<\/p>\n
In[6]:<\/h1>\nref: internet (not my code, using as a library)<\/h1>\n
def clean_dataset(df):
\nassert isinstance(df, pd.DataFrame), "df needs to be a pd.DataFrame"
\ndf.dropna(inplace=True)
\nindices_to_keep = ~df.isin([np.nan, np.inf, -np.inf]).any(1)
\nreturn df[indices_to_keep].astype(np.float64)<\/p>\n
In[7]:<\/h1>\nX has the features to cluster around (centres: Age, ACR). df has the complete data<\/h1>\nafter clustering is done using features in X, we find positions (index) for each data<\/h1>\nin a cluster then we use those index positions to cluster the data from df<\/h1>\n
after clustering is done using features in X, we find positions (index) for each data<\/h1>\nin a cluster then we use those index positions to cluster the data from df<\/h1>\n
X.shape, df.shape<\/p>\n
In[8]:<\/h1>\n
X = clean_dataset(X)<\/p>\n
In[9]:<\/h1>\ndefine the model<\/h1>\n
model = KMeans(n_clusters = 10) #,random_state=0, n_init="auto"<\/p>\n
fit the model<\/h1>\n
model.fit(X)
\n#model.labels_<\/p>\n
In[10]:<\/h1>\nCreate csv files with the cluster daya<\/h1>\nOne csv for one Cluster<\/h1>\nIn[11]:<\/h1>\n
One csv for one Cluster<\/h1>\nIn[11]:<\/h1>\n
howManyClusters = 10
\nfor clusterId in range (howManyClusters):
\nind_list = np.where(model.labels_ == clusterId )[0]
\ncluster = df.iloc[ind_list]
\ncluster.to_csv(‘.\/nhanes_output_data\/classifiedGroups\/kmeanscluster\/cluster-‘
\n+ str(clusterId) + ‘.csv’);<\/p>\n
In[12]:<\/h1>\n
model.cluster_centers_<\/p>\n
In[13]:<\/h1>\nScatter plot to see each cluster points visually<\/h1>\n
std_data = StandardScaler().fit_transform(X)
\nplt.scatter(std_data[:,0], std_data[:,1], c = model.labels_, cmap = "rainbow")
\nplt.title("K-means Clustering of Diet and ACR data")
\nplt.show()<\/p>\n
# References:<\/h1>\n# print("Shape of cluster:", model.cluster_centers_.shape)<\/h1>\n# https:\/\/stackoverflow.com\/questions\/50297142\/get-cluster-points-after-kmeans-in-a-list-format<\/a><\/h1>\n#<\/p>\n
# https:\/\/machinelearningmastery.com\/clustering-algorithms-with-python\/<\/a><\/h1>\n# https:\/\/stackoverflow.com\/questions\/50297142\/get-cluster-points-after-kmeans-in-a-list-format<\/a><\/h1>\n# https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.cluster.KMeans.html<\/a><\/h1>\n# https:\/\/datascience.stackexchange.com\/questions\/48693\/perform-k-means-clustering-over-multiple-columns<\/a><\/h1>\n# https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.cluster.KMeans.html<\/a><\/h1>\n#<\/p>\n
>>> from sklearn.cluster import KMeans<\/h1>\n>>> import numpy as np<\/h1>\n>>> X = np.array([[1, 2], [1, 4], [1, 0],<\/h1>\n… [10, 2], [10, 4], [10, 0]])<\/h1>\n>>> kmeans = KMeans(n_clusters=2, random_state=0, n_init="auto").fit(X)<\/h1>\n>>> kmeans.labels_<\/h1>\narray([1, 1, 1, 0, 0, 0], dtype=int32)<\/h1>\n>>> kmeans.predict([[0, 0], [12, 3]])<\/h1>\narray([1, 0], dtype=int32)<\/h1>\n>>> kmeans.cluster_centers_<\/h1>\narray([[10., \u00a02.],<\/h1>\n[ 1., \u00a02.]])<\/h1>\nIn[ ]:<\/h1>\n
# https:\/\/stackoverflow.com\/questions\/50297142\/get-cluster-points-after-kmeans-in-a-list-format<\/a><\/h1>\n#<\/p>\n
# https:\/\/machinelearningmastery.com\/clustering-algorithms-with-python\/<\/a><\/h1>\n# https:\/\/stackoverflow.com\/questions\/50297142\/get-cluster-points-after-kmeans-in-a-list-format<\/a><\/h1>\n# https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.cluster.KMeans.html<\/a><\/h1>\n# https:\/\/datascience.stackexchange.com\/questions\/48693\/perform-k-means-clustering-over-multiple-columns<\/a><\/h1>\n# https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.cluster.KMeans.html<\/a><\/h1>\n#<\/p>\n
>>> from sklearn.cluster import KMeans<\/h1>\n>>> import numpy as np<\/h1>\n>>> X = np.array([[1, 2], [1, 4], [1, 0],<\/h1>\n… [10, 2], [10, 4], [10, 0]])<\/h1>\n>>> kmeans = KMeans(n_clusters=2, random_state=0, n_init="auto").fit(X)<\/h1>\n>>> kmeans.labels_<\/h1>\narray([1, 1, 1, 0, 0, 0], dtype=int32)<\/h1>\n>>> kmeans.predict([[0, 0], [12, 3]])<\/h1>\narray([1, 0], dtype=int32)<\/h1>\n>>> kmeans.cluster_centers_<\/h1>\narray([[10., \u00a02.],<\/h1>\n[ 1., \u00a02.]])<\/h1>\nIn[ ]:<\/h1>\n
# https:\/\/stackoverflow.com\/questions\/50297142\/get-cluster-points-after-kmeans-in-a-list-format<\/a><\/h1>\n# https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.cluster.KMeans.html<\/a><\/h1>\n# https:\/\/datascience.stackexchange.com\/questions\/48693\/perform-k-means-clustering-over-multiple-columns<\/a><\/h1>\n# https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.cluster.KMeans.html<\/a><\/h1>\n#<\/p>\n
>>> from sklearn.cluster import KMeans<\/h1>\n>>> import numpy as np<\/h1>\n>>> X = np.array([[1, 2], [1, 4], [1, 0],<\/h1>\n… [10, 2], [10, 4], [10, 0]])<\/h1>\n>>> kmeans = KMeans(n_clusters=2, random_state=0, n_init="auto").fit(X)<\/h1>\n>>> kmeans.labels_<\/h1>\narray([1, 1, 1, 0, 0, 0], dtype=int32)<\/h1>\n>>> kmeans.predict([[0, 0], [12, 3]])<\/h1>\narray([1, 0], dtype=int32)<\/h1>\n>>> kmeans.cluster_centers_<\/h1>\narray([[10., \u00a02.],<\/h1>\n[ 1., \u00a02.]])<\/h1>\nIn[ ]:<\/h1>\n
# https:\/\/datascience.stackexchange.com\/questions\/48693\/perform-k-means-clustering-over-multiple-columns<\/a><\/h1>\n# https:\/\/scikit-learn.org\/stable\/modules\/generated\/sklearn.cluster.KMeans.html<\/a><\/h1>\n#<\/p>\n
>>> from sklearn.cluster import KMeans<\/h1>\n>>> import numpy as np<\/h1>\n>>> X = np.array([[1, 2], [1, 4], [1, 0],<\/h1>\n… [10, 2], [10, 4], [10, 0]])<\/h1>\n>>> kmeans = KMeans(n_clusters=2, random_state=0, n_init="auto").fit(X)<\/h1>\n>>> kmeans.labels_<\/h1>\narray([1, 1, 1, 0, 0, 0], dtype=int32)<\/h1>\n>>> kmeans.predict([[0, 0], [12, 3]])<\/h1>\narray([1, 0], dtype=int32)<\/h1>\n>>> kmeans.cluster_centers_<\/h1>\narray([[10., \u00a02.],<\/h1>\n[ 1., \u00a02.]])<\/h1>\nIn[ ]:<\/h1>\n
#<\/p>\n