This article shares another python interactive visualization tool Altair . The bottom layer of Altair is Vega-Lite (based on a simple interactive visualization grammar, A Grammar of Interactive Graphics), with effects such as:

The author of Altair is Jake Vanderplas. He is a big man. He was previously the Dean of the Institute of Physics of the University of Washington eScience School. He is now a Software Engineer at Google. He is passionate about Python, Astronomy and Data Science. He is also an active open source enthusiast. He can see his talk at PyData conferences over the years. In addition to Altair, he has made a lot of contributions to famous Python libraries such as Scikit-Learn, Scipy, Matplotlib, IPython, etc.; author of two high-star books Python Data Science Handbook and A Whirlwind Tour of Python.

The python interactive tools introduced earlier include pygal and cufflinks .

Python's Altair script is converted to JSON

2. Quick start of Altair complex graphics

configure_*() methods to personalize image attributes

Selection, condition, and binding make altair graphics better interact with the mouse

Layer, HConcat, VConcat, Repeat, Facet help altair easily build composite graphics

Chart.resolve_scale(), Chart.resolve_axis(), and Chart.resolve_legend() personalized composite graphics

3. Demo sharing based on Altair

A case on the official website about the weather

Official website: https://altair-viz.github.io/index.html

1. Quick start of Altair basic graphics

This summary introduces how to quickly draw common basic graphs , such as “bar”, “circle”, “square”, “tick”, “line”, * “area”, “point”, “rule”, “geoshape”, and "text" etc.

pip install altair

pip install altair vega_datasets -i https://pypi.tuna.tsinghua.edu.cn/simple#国内源加速安装

Altair drawing step by step

data preparation

The iris iris data set is still used. For the introduction of the data set, see: Python visualization|matplotlib10-drawing scatter plots

import seaborn as sns
pd_iris = sns.load_dataset("iris")
pd_iris.head(n=5)

Quick drawing

#快速绘图
import altair as alt
import pandas as pd

alt.Chart(pd_iris).mark_point().encode(x='sepal_length',
                                       y='sepal_width',
                                       color='species')

Drawing step split

From the code of alt. Chart (pd_iris). mark_ point(). encode (x='sepal_length',y='sepal_width',color='species'), we can see that Altair drawing mainly uses the Chart() method and mark_* () method, and encode() method.

The Chart() method converts the data into an altair.vegalite.v4.api.Chart object

The height, width, background color, etc. of the image can be set in the brackets. For details, see: https://altair-viz.github.io/user_guide/generated/toplevel/altair.Chart.html?highlight=chart

The mark_*() method specifies the graphics to be displayed, such as drawing a scatter chart mark_point()

The mark_*() method sets graphic attributes, such as color, size, etc.

Various attributes of the graphics to be displayed can be set in the brackets. Take mark_point() to set the point color as an example as follows.

For other detailed parameters, see: https://altair-viz.github.io/user_guide/generated/toplevel/altair.Chart.html?highlight=mark_point#altair.Chart.mark_point

The encode() method sets the mapping of the coordinate axis

Detailed parameters: https://altair-viz.github.io/user_guide/generated/toplevel/altair.Chart.html?highlight=encode#altair.Chart.encode

Python's Altair script is converted to JSON

python script

import altair as alt
import pandas as pd

data = pd.DataFrame({'x': ['A', 'B', 'C', 'D'], 'y': [1, 2, 1, 2]})
alt.Chart(data).mark_bar().encode(
    x='x',
    y='y',
)

Click to get the json script

{
  "config": {"view": {"continuousWidth": 400, "continuousHeight": 300}},
  "data": {"name": "data-39e740acccd9d827d4364cdbd6d37176"},
  "mark": "bar",
  "encoding": {
    "x": {"type": "nominal", "field": "x"},
    "y": {"type": "quantitative", "field": "y"}
  },
  "$schema": "https://vega.github.io/schema/vega-lite/v4.8.1.json",
  "datasets": {
    "data-39e740acccd9d827d4364cdbd6d37176": [
      {"x": "A", "y": 1},
      {"x": "B", "y": 2},
      {"x": "C", "y": 1},
      {"x": "D", "y": 2}
    ]
  }
}

2. Quick start of Altair complex graphics

This section briefly introduces more complex graphics, such as personalized facet diagram titles, legends, configure_*() methods , selection() binding_*() methods ,方法、condition()方法、

configure_*() methods to personalize image attributes

The configure_header() method personalizes the header

import altair as alt
from vega_datasets import data#vega_datasets为altair的一个内置数据集模块

source = data.cars.url

chart = alt.Chart(source).mark_point().encode(x='Horsepower:Q',
                                              y='Miles_per_Gallon:Q',
                                              color='Origin:N',
                                              column='Origin:N').properties(
                                                  width=180, height=180)
chart.configure_header(titleColor='green',
                       titleFontSize=14,
                       labelColor='red',
                       labelFontSize=14)

Personalize the legend of the configure_legend() method

import altair as alt
from vega_datasets import data

source = data.cars.url

chart = alt.Chart(source).mark_point().encode(x='Horsepower:Q',
                                              y='Miles_per_Gallon:Q',
                                              color='Origin:N')

chart.configure_legend(strokeColor='gray',
                       fillColor='#EEEEEE',
                       padding=10,
                       cornerRadius=10,
                       orient='top-right')

For more configure method introduction, please see: https://altair-viz.github.io/user_guide/configuration.html

Selection, condition, and binding make altair graphics better interact with the mouse

The selection(), condition(), and binding() methods are mainly used here , a brief introduction, see for details: https://altair-viz.github.io/user_guide/interactions.html

selection() method

The mouse can lightly capture a certain part of the graph.

condition() method

Make the part captured by the mouse brighten, and the part that is not captured is dimmed.

binding_*() method

The effect is as follows:

Layer, HConcat, VConcat, Repeat, Facet help altair easily build composite graphics

hconcat水平方向拼图

import altair as alt
from vega_datasets import data

iris = data.iris.url

chart1 = alt.Chart(iris).mark_point().encode(x='petalLength:Q',
                                             y='petalWidth:Q',
                                             color='species:N').properties(
                                                 height=300, width=300)

chart2 = alt.Chart(iris).mark_bar().encode(x='count()',
                                           y=alt.Y('petalWidth:Q',
                                                   bin=alt.Bin(maxbins=30)),
                                           color='species:N').properties(
                                               height=300, width=100)

chart1 | chart2

alt.hconcat(chart1, chart2)

vconcat vertical puzzle

LayerChart图层叠加

RepeatChart draws similar graphics

from vega_datasets import data

iris = data.iris.url

base = alt.Chart().mark_point().encode(color='species:N').properties(
    width=200, height=200).interactive()

chart = alt.vconcat(data=iris)
for y_encoding in ['petalLength:Q', 'petalWidth:Q']:
    row = alt.hconcat()
    for x_encoding in ['sepalLength:Q', 'sepalWidth:Q']:
        row |= base.encode(x=x_encoding, y=y_encoding)
    chart &= row
chart

FacetChart graphics facet

import altair as alt
from altair.expr import datum
from vega_datasets import data

iris = data.iris.url

base = alt.Chart(iris).mark_point().encode(x='petalLength:Q',
                                           y='petalWidth:Q',
                                           color='species:N').properties(
                                               width=160, height=160)

chart = alt.hconcat()
for species in ['setosa', 'versicolor', 'virginica']:
    chart |= base.transform_filter(datum.species == species)
chart

Chart.resolve_scale(), Chart.resolve_axis(), and Chart.resolve_legend() personalized composite graphics

For example, use resolve_scale() to use color wheels for two graphs respectively.

from vega_datasets import data

source = data.cars()

base = alt.Chart(source).mark_point().encode(
    x='Horsepower:Q', y='Miles_per_Gallon:Q').properties(width=200, height=200)

alt.concat(base.encode(color='Origin:N'),
           base.encode(color='Cylinders:O')).resolve_scale(color='independent')

3. Demo sharing based on Altair

A case of official gateway and weather

from vega_datasets import data

df = data.seattle_weather()
scale = alt.Scale(
    domain=['sun', 'fog', 'drizzle', 'rain', 'snow'],
    range=['#e7ba52', '#c7c7c7', '#aec7e8', '#1f77b4', '#9467bd'])
brush = alt.selection(type='interval')
points = alt.Chart().mark_point().encode(
    alt.X('temp_max:Q', title='Maximum Daily Temperature (C)'),
    alt.Y('temp_range:Q', title='Daily Temperature Range (C)'),
    color=alt.condition(brush,
                        'weather:N',
                        alt.value('lightgray'),
                        scale=scale),
    size=alt.Size('precipitation:Q',
                  scale=alt.Scale(range=[1, 200]))).transform_calculate(
                      "temp_range",
                      "datum.temp_max - datum.temp_min").properties(
                          width=600, height=400).add_selection(brush)

bars = alt.Chart().mark_bar().encode(
    x='count()',
    y='weather:N',
    color=alt.Color('weather:N', scale=scale),
).transform_calculate(
    "temp_range",
    "datum.temp_max - datum.temp_min").transform_filter(brush).properties(
        width=600)

alt.vconcat(points, bars, data=df)