pandas使用教程（十）：useful tricks！

一、create an example DataFrame

二、rename columns

df = df.rename({'col one':'col_one', 'col two':'col_two'}, axis='columns')

or

df.columns = ['col_one', 'col_two']

or

df.columns = df.columns.str.replace(' ', '_')

or

df.add_prefix('X_')   #df.add_suffix('_Y')

三、reverse row order

四、reverse column order

五、select columns by data type

drinks.select_dtypes(include='number').head()

or

drinks.select_dtypes(include='object').head()

or

drinks.select_dtypes(include=['object', 'number', 'category', 'datetime']).head()

or

drinks.select_dtypes(exclude='number').head()

六、convert strings to numbers

将NaN置0:

七、Reduce DataFrame size

pandas DataFrames are designed to fit into memory, so sometimes you need to reduce the DataFrame size

step1:只加载需要的columns：

step2:转换成category类型：

八、Build a DataFrame from multiple files(row-wise)

九、Build a DataFrame from multiple files(column-wise)

十、Create a DataFrame from a clipboard

step1:将表格中的内容复制

step2: run the codes

如果要复用，不建议使用read_clipboard()方法

十一、Split a DataFrame into two random subsets

十二、Filter a DataFrame by multiple categories

十三、Filter a DataFrame by largest categories

十四、Handle missing values

十五、Split a string into multiple columns

十六、Expand a Series of lists into a DataFrame

十七、 Reshape a MultiIndexed Series

十八、Create a pivot table

十九、Convert continuous(连续的) data into categorical(离散的) data I think it’s useful

二十、Change display options

二十一、Style a DataFrame

example1:

example2:

example3:

example4:

Bonus: Profile a DataFrame(查看一个DataFrame的简要概括)