How Can I Plot a Confusion Matrix? [Duplicate]

Question

I am using scikit-learn for classification of text documents(22000) to 100 classes. I use scikit-learn's confusion matrix method for computing the confusion matrix.

model1 = LogisticRegression()
model1 = model1.fit(matrix, labels)
pred = model1.predict(test_matrix)
cm=metrics.confusion_matrix(test_labels,pred)
print(cm)
plt.imshow(cm, cmap='binary')

This is how my confusion matrix looks like:

[[3962  325    0 ...,    0    0    0]
 [ 250 2765    0 ...,    0    0    0]
 [   2    8   17 ...,    0    0    0]
 ..., 
 [   1    6    0 ...,    5    0    0]
 [   1    1    0 ...,    0    0    0]
 [   9    0    0 ...,    0    0    9]]

However, I do not receive a clear or legible plot. Is there a better way to do this?

Accepted Answer

you can use plt.matshow() instead of plt.imshow() or you can use seaborn module's heatmap (see documentation) to plot the confusion matrix

import seaborn as sn
import pandas as pd
import matplotlib.pyplot as plt
array = [[33,2,0,0,0,0,0,0,0,1,3], 
        [3,31,0,0,0,0,0,0,0,0,0], 
        [0,4,41,0,0,0,0,0,0,0,1], 
        [0,1,0,30,0,6,0,0,0,0,1], 
        [0,0,0,0,38,10,0,0,0,0,0], 
        [0,0,0,3,1,39,0,0,0,0,4], 
        [0,2,2,0,4,1,31,0,0,0,2],
        [0,1,0,0,0,0,0,36,0,2,0], 
        [0,0,0,0,0,0,1,5,37,5,1], 
        [3,0,0,0,0,0,0,0,0,39,0], 
        [0,0,0,0,0,0,0,0,0,0,38]]
df_cm = pd.DataFrame(array, index = [i for i in "ABCDEFGHIJK"],
                  columns = [i for i in "ABCDEFGHIJK"])
plt.figure(figsize = (10,7))
sn.heatmap(df_cm, annot=True)

Answer 2

@bninopaul 's answer is not completely for beginners

here is the code you can "copy and run"

import seaborn as sn
import pandas as pd
import matplotlib.pyplot as plt

array = [[13,1,1,0,2,0],
         [3,9,6,0,1,0],
         [0,0,16,2,0,0],
         [0,0,0,13,0,0],
         [0,0,0,0,15,0],
         [0,0,1,0,0,15]]

df_cm = pd.DataFrame(array, range(6), range(6))
# plt.figure(figsize=(10,7))
sn.set(font_scale=1.4) # for label size
sn.heatmap(df_cm, annot=True, annot_kws={"size": 16}) # font size

plt.show()

Answer 3

IF you want more data in you confusion matrix, including "totals column" and "totals line", and percents (%) in each cell, like matlab default (see image below)

including the Heatmap and other options...

You should have fun with the module above, shared in the github ; )

This module can do your task easily and produces the output above with a lot of params to customize your CM:

How Can I Plot a Confusion Matrix? [Duplicate]

Must Read

3 Answers

David Miller