cols = df.columns.tolist()<br /> # Rearrange the list any way you want<br /> cols = cols[-1:] + cols[:-1]<br /> df = df[cols]

# setting up a dummy dataframe<br /> raw_data = {‘name’: [‘Willard Morris’, ‘Al Jennings’, ‘Omar Mullins’, ‘Spencer McDaniel’],<br /> ‘age’: [20, 19, 22, 21],<br /> ‘favorite_color’: [‘blue’, ‘red’, ‘yellow’, “green”],<br /> ‘grade’: [88, 92, 95, 70]}<br /> df = pd.DataFrame(raw_data, index = [‘Willard Morris’, ‘Al Jennings’, ‘Omar Mullins’, ‘Spencer McDaniel’])<br /> df</p> <p>#now ‘age’ will appear at the end of our df<br /> df = df[[‘favorite_color’,’grade’,’name’,’age’]]<br /> df.head()

#old df columns<br /> df.columns<br /> Index([‘A’, ‘B’, ‘C’, ‘D’],dtype=”***”)<br /> #new column format that we want to rearange<br /> new_col = [‘D’,’C’,’B’,’A’] #list of column name in order that we want</p> <p>df = df[new_col]<br /> df.columns<br /> Index([‘D’, ‘C’, ‘B’, ‘A’],dtype=”***”)<br /> #new column order

# Get column list in [‘item1′,’item2′,’item3’] format<br /> df.columns<br /> # [0]output:<br /> Index([‘item1′,’item2′,’item3’], dtype=”object”)</p> <p># Copy just the list portion of the output and rearrange the columns<br /> cols = [‘item3′,’item1′,’item2’]</p> <p># Resave dataframe using new column order<br /> df = df[cols]<br />

df = df[[‘col1’, ‘col2’, ‘col3’, ‘col4’]]

You could also do something like this:</p> <p>df = df[[‘mean’, ‘0’, ‘1’, ‘2’, ‘3’]]<br /> You can get the list of columns with:</p> <p>cols = list(df.columns.values)<br /> The output will produce:</p> <p>[‘0’, ‘1’, ‘2’, ‘3’, ‘mean’]

Reorder columns in pandas