How do I show a scatter plot in Python after doing PCA?

Question

I made a random data of my own, that comprises of a text file with 18 rows and 5 columns with all integer entries.

I successfully managed to do PCA but now stuck. I am unable to do a scatter plot. Here is my code:

f=open(r'<path>mydata.txt')
print(f.read()) #reading from a file


with open(r'<path>mydata.txt') as f:
emp= []
for line in f:
    line = line.split() 
    if line:            
        line = [int(i) for i in line]
        emp.append(line)


from sklearn.decomposition import PCA
import pylab as pl
from itertools import cycle
X = emp
pca = PCA(n_components=3, whiten=True).fit(X)
X_pca = pca.transform(X) #regular PCA

Now, with PCA done and my variances known, how do I plot?

Here is how a sample data in my data set looks:

2    1    2    3    0
2    3    2    3    0
1    3    1    1    0
1    5    2    1    0
2    3    1    1    0
3    3    0    1    0
7    1    1    1    1
7    2    2    1    1
1    1    1    4    1
3    2    3    2    1
2    2    2    2    1
1    3    2    3    1
2    3    2    1    2
2    2    1    1    2
7    5    3    2    2
3    4    2    4    2
2    1    1    1    2
7    1    3    3    2

Answer 1

Is this what you are asking for?

import numpy as np
from matplotlib import pyplot as plt


data1 = [np.random.normal(0,0.1, 10), np.random.normal(0,0.1,10)]
data2 = [np.random.normal(1,0.2, 10), np.random.normal(2,0.3,10)]
data3 = [np.random.normal(-2,0.1, 10), np.random.normal(1,0.5,10)]


plt.scatter(data1[0],data1[1])
plt.scatter(data2[0],data2[1])
plt.scatter(data3[0],data3[1])

plt.show()

the result for the three different data sets would look something like this:

EDIT:

Hopefully I now understand your question better. Here the new code:

import numpy as np
from matplotlib import pyplot as plt    

with open(r'mydata.txt') as f:
    emp= []
    for line in f:
        line = line.split() 
        if line:            
            line = [int(i) for i in line]
            emp.append(line)


from sklearn.decomposition import PCA
import pylab as pl
from itertools import cycle
X = emp
pca = PCA(n_components=3, whiten=True).fit(X)
X_pca = pca.transform(X) #regular PCA

jobs = ['A', 'B', 'C']
job_id = np.array([e[4] for e in emp])

fig, axes = plt.subplots(3,3, figsize=(5,5))

for row in range(axes.shape[0]):
    for col in range(axes.shape[1]):
        ax = axes[row,col]
        if row == col:
            ax.tick_params(
                axis='both',which='both',
                bottom='off',top='off',
                labelbottom='off',
                left='off',right='off',
                labelleft='off'
            )
            ax.text(0.5,0.5,jobs[row],horizontalalignment='center')
        else:
            ax.scatter(X_pca[:,row][job_id==0],X_pca[:,col][job_id==0],c='r')
            ax.scatter(X_pca[:,row][job_id==1],X_pca[:,col][job_id==1],c='g')
            ax.scatter(X_pca[:,row][job_id==2],X_pca[:,col][job_id==2],c='b')
fig.tight_layout()
plt.show()

I named the jobs 'A', 'B', and 'C' with the ids 0, 1, and 2, respectively. From the last row of emp, I create a numpy array that holds these indices. In the crucial plotting commands, I mask the data by the job ids. Hope this helps.

The resulting plot looks like this:

EDIT 2:

If you want only one plot where you correlate, say, the first and the second column of X_pca with each other, the code becomes much more simple:

import numpy as np
from matplotlib import pyplot as plt

with open(r'mydata.txt') as f:
    emp= []
    for line in f:
        line = line.split() 
        if line:            
            line = [int(i) for i in line]
            emp.append(line)


from sklearn.decomposition import PCA
import pylab as pl
from itertools import cycle
X = emp
pca = PCA(n_components=3, whiten=True).fit(X)
X_pca = pca.transform(X) #regular PCA

jobs = ['A', 'B', 'C']
job_id = np.array([e[4] for e in emp])

row = 0
col = 1

plt.scatter(X_pca[:,row][job_id==0],X_pca[:,col][job_id==0],c='r')
plt.scatter(X_pca[:,row][job_id==1],X_pca[:,col][job_id==1],c='g')
plt.scatter(X_pca[:,row][job_id==2],X_pca[:,col][job_id==2],c='b')

plt.show()

The result looks like this:

I strongly suggest that you read the documentation of the functions used in these examples.

Answer 2

Based on your comment that you want to get this (https://i.stack.imgur.com/VsicE.jpg), here is how to do it using sklearn library:

In this example I am using the iris data:

PART 1: Plot only the scatter plot

import numpy as np
import matplotlib.pyplot as plt
from sklearn import datasets
from sklearn.decomposition import PCA
from numpy import linalg as LA
import pandas as pd
from scipy import stats

iris = datasets.load_iris()
X = iris.data
y = iris.target
#In general a good idea is to scale the data
X = stats.zscore(X)

pca = PCA()
x_new = pca.fit_transform(X)

plt.scatter(x_new[:,0], x_new[:,1], c = y)
plt.xlabel('PC1')
plt.ylabel('PC2')
plt.show()

RESULT 1

PART 2: in case you want to plot the famous biplot

#Create the biplot function
def biplot(score,coeff,labels=None):
    xs = score[:,0]
    ys = score[:,1]
    n = coeff.shape[0]
    scalex = 1.0/(xs.max() - xs.min())
    scaley = 1.0/(ys.max() - ys.min())
    plt.scatter(xs * scalex,ys * scaley, c = y)
    for i in range(n):
        plt.arrow(0, 0, coeff[i,0], coeff[i,1],color = 'r',alpha = 0.5)
        if labels is None:
            plt.text(coeff[i,0]* 1.15, coeff[i,1] * 1.15, "Var"+str(i+1), color = 'g', ha = 'center', va = 'center')
        else:
            plt.text(coeff[i,0]* 1.15, coeff[i,1] * 1.15, labels[i], color = 'g', ha = 'center', va = 'center')
plt.xlim(-1,1)
plt.ylim(-1,1)
plt.xlabel("PC{}".format(1))
plt.ylabel("PC{}".format(2))
plt.grid()


#Call the function. Use only the 2 PCs.
biplot(x_new[:,0:2],np.transpose(pca.components_[0:2, :]))
plt.show()

RESULT 2