分组数据并可视化缺失值

Question

I do have a large medical dataset that I want to group by hospital then plot graph of missing values per hospital. Here is how the dataset looks like:

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

# intialise data of lists. 
data = {'hospital':['Nick hospital', 'Nick hospital','Nick hospital', 'Krish hospital', 'Krish hospital','Krish hospital'], 
        'NAR_used':[2, 1,np.nan, 2, np.nan,3], 'ipno':[45,np.nan,np.nan,np.nan,65,67]
       } 

# Create DataFrame 
df = pd.DataFrame(data)
df

With this sample dataset, we only have 2 hospitals on hospital variable hence I want to visualize missing values for each hospital. Here is what I tried out

grouped = df.groupby(['hospital'])
for (i in grouped):
    null_counts = df.isnull().sum()/len(df)
    plt.figure(figsize=(16,8))
    plt.xticks(np.arange(len(null_counts)) + 0.5, null_counts.index, 
    rotation = 'vertical')
    plt.ylabel('Fraction of rows with missing data')
    plt.bar(np.arange(len(null_counts)), null_counts)

My solution is not generating graphs for each Hospital. Kindly help. Expected output is visualizing a bar graph of missing values for each hospital .

Answer 1

Let's try:

df.set_index('hospital').isna().sum(level=0).plot.bar()

Output: