Is there a more efficient way to correspond values from different tables?

Asked Mar 10 '19 at 19:19

Active Mar 10 '19 at 19:21

Viewed 25 times

Can I run this code with more efficiency? This is a simplified example, I'm actually using a table with almost a million entries and doing multiple loops. It's taking a long time to run. Any advice is welcome!

import pandas as pd

codeTable = {'Code': [1,2,3], 'Activity': ['a1', 'a2', 'a3']}
Table = {'Code':[1,2,4,3,6,7,15,2,3,4]}

df1 = pd.DataFrame(data = codeTable)
df2 = pd.DataFrame(data = Table)
df2['Activity'] = "-"
for i in range (0,df2.shape[0]):
    if df2['Code'][i] in set(df1['Code']):
        df2['Activity'][i] = df1['Activity'][ df1['Activity'][df1['Code'] == df2['Code'][i] ].index [0] ]
    else:
        df2['Activity'][i] = "-"

edited Mar 10 '19 at 19:21

asked Mar 10 '19 at 19:19

Luan Vieira

1

Use `df2['Activity'] = df2['Code'].map(df1.set_index('Code')['Activity']).fillna('-')` – jezrael Mar 10 '19 at 19:39
1

Thank you! Exactly what I want. – Luan Vieira Mar 10 '19 at 19:59

Is there a more efficient way to correspond values from different tables?

0 Answers0