pandas 笔记：高亮内容

1 高亮缺失值

1.0 数据

import pandas as pd
import numpy as np

data=[{'a':1,'b':2},
     {'a':3,'c':4},
     {'a':10,'b':-2,'c':5}]

df1=pd.DataFrame(data)
df1

 1.1 highlight_null

df.style.highlight_null(
    color: 'str' = 'red',
    subset: 'Subset | None' = None,
    props: 'str | None' = None,
)

 1.1.1 默认情况

df1.style.highlight_null()

1.1.2  color

调整高亮的颜色（后同）

df1.style.highlight_null(color='green')

 1.1.3 subset

指定操作的列（后同）

df1.style.highlight_null(subset='b')

1.1.4 props

 突出显示的CSS属性（后同）

df1.style.highlight_null(props='color:pink; background-color:blue')

 

 2 高亮最大/最小值

2.0 数据

# Visual Python: Data Analysis > File
vp_df = pd.read_csv('https://raw.githubusercontent.com/visualpython/visualpython/main/visualpython/data/sample_csv/iris.csv')
vp_df=vp_df[:5]
vp_df

 2.1 highlight_max

df.style.highlight_max(
    subset: 'Subset | None' = None,
    color: 'str' = 'yellow',
    axis: 'Axis | None' = 0,
    props: 'str | None' = None,
)

subset color props 同1.1

2.1.1 axis

默认为0，表示列

axis=1，表示行

vp_df.style.highlight_max(axis=1,
                          subset=['sepal_length','sepal_width','petal_length','petal_width'])

 axis=None, dataFrame中最大的那个

 2.2 highlight_min

参数和使用方法和max一样

2.2.1 链式调用min和max

vp_df.style.highlight_max(color='purple').highlight_min(color='green')

 3 高亮区间值

还是使用vp_df

df.style.highlight_between(
    subset: 'Subset | None' = None,
    color: 'str' = 'yellow',
    axis: 'Axis | None' = 0,
    left: 'Scalar | Sequence | None' = None,
    right: 'Scalar | Sequence | None' = None,
    inclusive: 'str' = 'both',
    props: 'str | None' = None,
)

subset、color、axis、props和前面的min、max一样

3.1 left、right

指定区间最小值和最大值

vp_df.style.highlight_between(left=3,
                              right=4.9,
                              subset=['sepal_length','sepal_width','petal_length','petal_width'])

3.2  inclusive 

用于确定是否左右闭包，可选'both', 'neither', 'left', 'right' （保留哪个）

vp_df.style.highlight_between(left=3,
                              right=4.9,
                              inclusive='both',
                              subset=['sepal_length','sepal_width','petal_length','petal_width'])

vp_df.style.highlight_between(left=3,
                              right=4.9,
                              inclusive='left',
                              subset=['sepal_length','sepal_width','petal_length','petal_width'])

vp_df.style.highlight_between(left=3,
                              right=4.9,
                              inclusive='right',
                              subset=['sepal_length','sepal_width','petal_length','petal_width'])

vp_df.style.highlight_between(left=3,
                              right=4.9,
                              inclusive='neither',
                              subset=['sepal_length','sepal_width','petal_length','petal_width'])

 3.3 不同的列不同的区间

vp_df.style.highlight_between(subset=['sepal_length','sepal_width','petal_length','petal_width'],
                              left=[4.6,3.1,1.33,0.2],
                              right=[4.9,3.5,1.5,0.3],
                              axis=1,
                              color='green'
                              )

 4 高亮分位数

df.style.highlight_quantile(
    subset: 'Subset | None' = None,
    color: 'str' = 'yellow',
    axis: 'Axis | None' = 0,
    q_left: 'float' = 0.0,
    q_right: 'float' = 1.0,
    interpolation: 'str' = 'linear',
    inclusive: 'str' = 'both',
    props: 'str | None' = None,
) 

subset、color、axis、inclusive、props和之前的一样

4.1  q_left、q_right

用于指定分位数左边界和右边界

vp_df.style.highlight_quantile(subset=['sepal_length','sepal_width','petal_length','petal_width'],
                              q_left=0.1,
                              q_right=0.8,
                              color='green'
                              )

 5 背景渐变色

df.style.background_gradient(
    cmap='PuBu',
    low: 'float' = 0,
    high: 'float' = 0,
    axis: 'Axis | None' = 0,
    subset: 'Subset | None' = None,
    text_color_threshold: 'float' = 0.408,
    vmin: 'float | None' = None,
    vmax: 'float | None' = None,
    gmap: 'Sequence | None' = None,
) 

cmap
low、high	指定最小最大值颜色边界 区间为[0,1]
axis	指定行、列或全部
subset	指定操作的列或行
text_color_threshold	指定文本颜色亮度，区间[0, 1]
vmin,vmax	指定与cmap最小最大值对应的单元格最小最大值

vp_df.style.background_gradient()

vp_df.style.background_gradient(vmin=3,vmax=4.9)

6 文本渐变色 

vp_df.style.text_gradient(cmap='RdYlGn')

 7 数据条

df.style.bar(
    subset: 'Subset | None' = None,
    axis: 'Axis | None' = 0,
    color='#d65f5f',
    width: 'float' = 100,
    align: 'str' = 'left',
    vmin: 'float | None' = None,
    vmax: 'float | None' = None,
)

axis	指定行、列或全部
subset	指定操作的列或行
color	数据条颜色
width	指定数据条长度
vmin、vmax	指定与数据条最小最大值对应的单元格最小最大值
align	数据条与单元格对齐方式，默认是left左对齐，还有zero居中和mid位于（max-min）/2

vp_df.style.bar(subset='sepal_length',
                color='green',
                vmin=4.5,
                vmax=5.1,
                align='left').bar(subset='petal_length',
                color='blue',
                vmin=1.3,
                vmax=1.5,
                align='zero',
                width=10)

8 逐元素条件着色

# Visual Python: Data Analysis > File
import pandas as pd
vp_df = pd.read_csv('https://raw.githubusercontent.com/visualpython/visualpython/main/visualpython/data/sample_csv/iris.csv')
vp_df=vp_df[:5][['sepal_length','sepal_width','petal_length','petal_width']]
vp_df

 

def color_red(val):
    color = 'purple' if val < 3.5 else 'green'
    return 'color: %s' % color
vp_df.style.applymap(color_red)