The Analysis Workflow | Pychallenger

Each previous lesson covered one step: exploring, counting, deriving, visualizing. In practice, these steps chain together into a repeatable workflow.

Here's that workflow on a small sales dataset. This one has missing values, unlike the clean companies data.

Step 1: Size up the data (familiar from the first lesson):

import pandas as pd
 
df = pd.DataFrame({
    'Name': ['A','B','C','D','E'],
    'Sales': [100,None,300,150,None],
    'Region': ['N','S','N','S','N']
})
print(df.shape)

Python

Output

Step 2: Check for gaps. Two Sales entries are missing:

print(df.isnull().sum())

Python

Output

What will be the output?

import pandas as pd
df = pd.DataFrame({
    'A': [1, None, 3],
    'B': [None, None, 6]
})
r = df.isnull().sum().tolist()
print(r)

Python

Step 3: Fill the gaps. Replace missing Sales with the column average:

mean_sales = df['Sales'].mean()
df['Sales'] = df['Sales'].fillna(
    mean_sales
)
print(df['Sales'].tolist())

Python

Output

What will be the output?

import pandas as pd
s = pd.Series([10, None, 30])
s = s.fillna(s.mean())
print(s.tolist())

Python

Step 4: Answer a question. Which region sells more on average?

result = df.groupby('Region')
result = result['Sales'].mean()
print(result)

Python

Output

What will be the output?

import pandas as pd
df = pd.DataFrame({
    'G': ['X', 'X', 'Y'],
    'V': [10, 30, 50]
})
r = df.groupby('G')['V'].mean()
print(r['X'])

Python

Step 5: Make the answer visual:

import matplotlib.pyplot as plt
 
result = df.groupby('Region')
result = result['Sales'].mean()
 
plt.bar(result.index, result.values)
plt.xlabel('Region')
plt.ylabel('Mean Sales')
plt.title('Sales by Region')
plt.show()

Python

Five steps: explore → clean → analyze → visualize → answer. Each feeds into the next.

Real workflows loop back. A chart might reveal outliers that require more cleaning, or spark an entirely new question.

What will be the output?

import pandas as pd
df = pd.DataFrame({
    'A': [1, None, 3, 4],
    'B': ['x', 'y', 'z', 'w']
})
print(df.shape, df['A'].dtype)

Python