import pandas as pd import numpy as np import matplotlib.pyplot as plt import seaborn as sns from pylab import rcParams import missingno as msno import warnings # Ignores any warning rcParams['figure.figsize'] = 12, 5 pd.set_option('display.max_rows', None) pd.set_option('display.max_columns', None) pd.set_option ('display.large_repr', 'truncate') pd.set_option ('display.max_colwidth', 30) #Here the first rows of the dataset which includes the unwanted headers are removed. train = pd.read_csv(r"C:\\Users\\data analysis\\sales data correct .csv", header = 6) #If the country is empty such rows are omitted. unwantedrows = train.dropna (subset = ["Country"], inplace = True) print (train.head(5)) #to plot msno.matrix(train) plt.show() I have a large dataset with missing values. To find the pattern of missing values I am trying to visualize them according to the column values. The dataset contains quality, country and a couple of years. It is about sales occurred during these years. Manually it can be identified that missing values are because the company has not introduced themselves to those countries in the missing years. How to find this visually using any python libraries? What I need is in Y axis the Quality column values (Standard, premium) to be displayed and x axis contains the rest of columns including quality. Quality Country 1994 1995 1996 Standard USA 123 345 789 Standard UK NA 45 78 Standard Europe NA NA 56 Premium USA 56 67 56 Premium UK NA 45 24 Sample of what I need Tried using msno.matrix. But it does not state the column values. Continue reading...
