Pokemon Data Analyzing in Python Using Machine Learning

By Sudipta Ghosh

The project aims to see the mean value of Pokemon attack and defense and also plot a graph between attack and defense

Step: 1

First, we will import the necessary libraries then we upload the dataset

import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt

P = pd.read_csv('Pokemon_A.csv')
P

P.info()

Step: 2

Now we see agglomeration values

Q=P[['Attack','Defence']]

def doAgglomerative(X, nclust=2):
    Model = AgglomerativeClustering(n_clusters=nclust,affinity='euclidean',linkage='ward')
    Clust_Labels = Model.fit_predict(X)
    return (Clust_Labels)

Clust_Labels = doAgglomerative(Q,20)
agglomerative = pd.DataFrame(Clust_Labels)
Q.insert((Q.shape[1]),'agglomerative',agglomerative)

T=Q
T['Name']=P['Name']
T

for i in range(20):
    names0=T[T['agglomerative']==i]['Name'] 
    print('[ Agglomerative '+str(i)+' ]')
    print(list(set(names0)))
    print()

Step: 3

Now we calculate the mean value of 

Q.groupby('agglomerative').mean()

Step: 4

Now we plot a graph of Agglomerative Clustering between Attack and Defenses

fig = plt.figure()
ax = fig.add_subplot(111)
scatter = ax.scatter(Q['Attack'],Q['Defence'],c=agglomerative[0],s=50)
ax.set_title('Agglomerative Clustering')
ax.set_xlabel('Attack')
ax.set_ylabel('Defence')
plt.rcParams['figure.figsize'] = (6,5)
plt.colorbar(scatter)