Fill missing data, categorical & numerical

import pandas as pd
import numpy as np

# Create simple dictionary with some missing values
data = {
    'Name': ['John', 'Anna', np.nan, 'Linda', 'John'],
    'Type': ['Type1', 'Type2', 'Type2', np.nan, 'Type1'],
    'Country': ['Country1', np.nan, 'Country2', 'Country1', 'Country2'],
}

# Convert dictionary to pandas DataFrame
data = pd.DataFrame(data)

print("Original DataFrame:")
print(data)

# Apply mode and fillna
for column in ['Name', 'Type', 'Country']:
    mode = data[column].mode()[0] # in case we have more then one mode
    data[column].fillna(mode, inplace=True)

print("\nDataFrame after filling NA values with mode:")
print(data)

# Original DataFrame:

    Name   Type   Country
0   John  Type1  Country1
1   Anna  Type2       NaN
2    NaN  Type2  Country2
3  Linda    NaN  Country1
4   John  Type1  Country2



# DataFrame after filling NA values with mode:

    Name   Type   Country
0   John  Type1  Country1
1   Anna  Type2  Country1
2   John  Type2  Country2
3  Linda  Type1  Country1
4   John  Type1  Country2

import pandas as pd
import numpy as np

# Sample data with some missing values
data = {
    'Turnover': [100, 200, np.nan, 400, 500, 600, np.nan, 800],
    'Transactions': [1, 2, 3, np.nan, np.nan, 6, 7, 8]
}

# Convert dictionary to a pandas DataFrame
data = pd.DataFrame(data)

print("Original DataFrame:")
print(data)

# Fill missing values with column median
for column in ['Turnover', 'Transactions']:
    median = data[column].median()
    data[column].fillna(median, inplace=True)

print("\nDataFrame after filling NA values with median:")
print(data)

# Original DataFrame:

   Turnover  Transactions
0     100.0           1.0
1     200.0           2.0
2       NaN           3.0
3     400.0           NaN
4     500.0           NaN
5     600.0           6.0
6       NaN           7.0
7     800.0           8.0



# DataFrame after filling NA values with median:

   Turnover  Transactions
0     100.0           1.0
1     200.0           2.0
2     450.0           3.0
3     400.0           4.5
4     500.0           4.5
5     600.0           6.0
6     450.0           7.0
7     800.0           8.0

import pandas as pd
import numpy as np

# Simple DataFrame with both numerical and categorical data
data = {
    'Age': [25, 30, 35, np.nan, 45],
    'City': ['New York', 'Seattle', 'San Francisco', 'Austin', np.nan],
    'Income': [50000, 70000, np.nan, 90000, 100000]
}

df = pd.DataFrame(data)
print("Original DataFrame:")
print(df)

def fill_na_with_mean(df, num_cols):
    mean_values = df[num_cols].mean()
    df[num_cols] = df[num_cols].fillna(mean_values)
    return df


def fill_na_with_mode(df, cat_cols):
    mode_values = df[cat_cols].mode().iloc[0]
    df[cat_cols] = df[cat_cols].fillna(mode_values)
    return df


def clean_data(df):
    # Identify numerical columns and fill NA/NaN with mean
    num_cols = df.select_dtypes(include=np.number).columns
    df = fill_na_with_mean(df, num_cols)

    # Identify categorical columns and fill NA/NaN with mode
    cat_cols = df.select_dtypes(include='object').columns
    df = fill_na_with_mode(df, cat_cols)

    return df



# Apply the function on our DataFrame
cleaned_df = clean_data(df)
print("\nCleaned DataFrame:")
print(cleaned_df)

# Original DataFrame:

    Age           City    Income
0  25.0       New York   50000.0
1  30.0        Seattle   70000.0
2  35.0  San Francisco       NaN
3   NaN         Austin   90000.0
4  45.0            NaN  100000.0


# Cleaned DataFrame:

     Age           City    Income
0  25.00       New York   50000.0
1  30.00        Seattle   70000.0
2  35.00  San Francisco   77500.0
3  33.75         Austin   90000.0
4  45.00         Austin  100000.0