Python Pandas cheat sheet

Python Pandas Library Cheat Sheet for a Beginner

This is a python pandas cheat sheet for a quick lookup and review. For more information and explanation, please visit pandas doc: https://pandas.pydata.org/docs/reference/frame.html

Loading data:

1. Import: Import pandas as pd

2. Read CSV: pd.read_csv(‘file_name.csv’)

3. Union all multiple dataframes: pd.concat([df1, df2, df3, …])

4. Merge:

5. pd.merge(table1, table2)

6. table1.merge(table2)

7. Using left_on or right_on: pd.merge(table1, table2, left_on: table1_column, right_on: table2_column)

8. Useful option: suffixes = [table1_suffix, table2_suffix]

9. Join types: pd.merge(table1, table2, how=’join_type’)

10. Join types = [‘left’, ‘right’, ‘outer’]  

11. Default join type = ‘inner’

Inspecting data:

1. Inspection: 

2. dataframe.columns

3. dataframe.head()

4. dataframe.describe()  

5. dataframe.info()

6. dataframe.dtypes

7. dataframe.isnull().sum() << shows you the # of rows per column with null value

8. Duplicates:

9. Finding duplicates: dataframe.duplicated() 

10. Removing duplicates: dataframe.drop_duplicates()

11. Getting the column information using aggregate: 

12. Available: sum(), mean(), median(), max(), min(), std(), nunique(), unique()

13. dataframe.column_name.aggregate_function(): ex) dataframe.column_name.sum()

14. Group by:

15. One groupby: dataframe.groupby(‘column1’).column2.count()

16. Multiple groupby: dataframe.groupby([‘column1’, ‘column2’]).column3.count()

17. Selecting columns: 

18. One column: 

19. Dataframe[‘column1’]

20. If the column name is one word: dataframe.column1  

21. Multiple columns: dataframe[[‘column1’,’column2’,’column3’]]

Looking up:

1. Filtering dataframe using loc or iloc:

2. loc: using label. EX) dataframe.loc[]

3. iloc: using index. EX) dataframe.iloc[]

4. Getting the result for specific values: 

5. One: dataframe[dataframe.column_name == ‘specific value]

6. You can use other logical operations like: <, >, != 

7. Multiple: use format of: 

8. And: () & ()

9. Or: () | ()

Cleaning and Tidying data:

1. Column name change: dataframe.rename(columns = {‘old_column_name’:’new_column_name’, ‘old_column_name2’:’new_column_name2’})

2. Create a new column: 

3. New_column_value can be one value: ex)  dataframe[‘new_column_name’] = ‘Yes’

4. New_column_value can be a simple calc: ex)  dataframe[‘new_column_name’] = dataframe[column1] * dataframe[column2]

5. New_column_value can be more complex calc using lambda: ex) dataframe[‘new_column_name’] = dataframe.apply( lambda x: x[‘column1’]**2 if x[‘column2’] > (x[‘column1’]*2) else x[‘column2’]*3)

6. Dealing with missing data: 

7. Drop them: dataframe.dropna()

8. Fill them: 

9. One column: dataframe.column.fillna(value)

10. Multiple columns: dataframe.fillna(value = {column1: value1, column2:value2})

11. Pivot: 

12. dataframe.pivot(index = , columns = , values = ).reset_index()

13. pd.melt(dataframe, id_vars = , var_name = , value_name = )

Export:

1. Export: 

2. To csv: dataframe.to_csv()

3. To json: dataframe.to_json()

#cheatsheet #cheatsheet #pandas #python