pandas_cheatsheet.md

	type	size	ratio
1	A	3	0.2
2	B	5	0.4
3	C	NaN	0.8

Pandas Data Structures

Create a DataFrame

>>> import pandas as pd

>>> df = pd.DataFrame({'type': ['A', 'B', 'C'], 'size': [3, 5, None], 'ratio': [0.2, 0.4, 0.8]})
>>> df
  type  size  ratio
0    A   3.0    0.2
1    B   5.0    0.4
2    C   NaN    0.8

I/O

Read and Write to CSV

>>> df = pd.read_csv('file.csv', header=None, nrows=5)
>>> df.to_csv('myDataFrame.csv')

Selection

Getting

df['type']  # Select all values from type column
df[['type', 'size']]  # Select all values from type and size columns

Selecting', Boolean Indexing and Setting

By Position

df.iloc[:2, 0]

By label

df.loc[:2, 'type']

Boolean Indexing

df[df['ratio'] > 0.5]
df[(df['ratio'] > 0.5) & (df['type'] == 'A')]  # or |, and &, not ~

Setting, create a new column

df['ratio_x_100'] = df['ratio'] * 100

Dropping

columns

df.drop(columns=['type'])
df.drop('type', axis=1)

missing values

df.dropna()

Sort

indexs

df.sort_index()

values

df.sort_values(by='Country') 
df['size']

Retrieving Series/DataFrame Information

basic info

df.shape
df.index
df.columns
df.dtypes

Summary statistics

df.describe()

number of valid values

df.count()

Summary

df.sum()
df.cumsum()
df.min()
df.max()
df.mean()
df.median()
df.std()

Applying functions

df.apply(lambda x: x + 2, axis=0)

Arithmetic Operations

ss = df['size'] * df['ratio']

Fill Methods

df['size'].fillna(0)
df['size'].fillna()

Plotting

df['size'].plot()
df.plot.hist(bins=10)