How to filter a groupby sum-CodePudding

I am looking to total sections of a total column using the groupby function. When I use the groupby function 'code' it works however I would like to be able to filter it down to one nominal code by placing it in a variable and printing it.

subheading_one = df.groupby(['Code'])['Total'].sum()
subheading_two = df.groupby(['Code'])['Total'].sum()
subheading_three = df.groupby(['Code'])['Total'].sum()

print('Cost heading 1.1 £: ',subheading_one)
print('Cost heading 1.2 £: 'subheading_two)
print('Cost heading 1.3 £: 'subheading_three)

I have attached a snippet of the data frame. As you can see I would like to be able to total only '1.1' items and place that total into a variable (should equal 300). Can anybody help?

CodePudding user response：

You should be able to perform the .groupby operation once, and then use .loc to select the specific total you want:

totals = df.groupby(['Code'])['Total'].sum()

print('Cost heading 1.1 £: ', totals.loc['1.1'])
print('Cost heading 1.2 £: ', totals.loc['1.2'])

edit: If your code column are floats (not strings), you can do:

totals = df.groupby(['Code'])['Total'].sum()

print('Cost heading 1.1 £: ', totals.loc[1.1])
print('Cost heading 1.2 £: ', totals.loc[1.2])

CodePudding user response：

You can use the get_group like this

subheading_one = df.groupby(['Code'])['Total'].get_group('1.1').sum()

I am assuming your Code feature has string values.

CodePudding user response：

    code = {'Code': [1.1, 1.1, 1.1, 1.2, 1.2, 1.3, 1.4, 1.5]}
    df = pd.DataFrame(code)
    df['Description'] = 'Item here'
    df['Quantity'] = 1
    df['Unit'] = 'Item'
    df['Rate'] = 100
    df['Total'] = 100

    mask = df['Code'] == 1.1
    subheading_one = df.loc[mask, 'Total'].sum()
    mask = df['Code'] == 1.2
    subheading_two = df.loc[mask, 'Total'].sum()
    mask = df['Code'] == 1.3
    subheading_three = df.loc[mask, 'Total'].sum()

    print(f'Cost heading 1.1 £: {subheading_one}')
    print(f'Cost heading 1.2 £: {subheading_two}')
    print(f'Cost heading 1.3 £: {subheading_three}')