Welcome back to my 100 Days of Data Science Challenge Journey. On day 6, I worked on Bar Chart Race Animation using Python. For today’s task, I will be using the Countries Population from 1955 to 2020 dataset available on Kaggle.

欢迎回到我的100天数据科学挑战之旅。 第6天，我使用Python制作了条形图竞赛动画。 对于今天的任务，我将使用Kaggle上可用的1955年至2020年的国家人口数据集。

You can read my previous days stories here:

您可以在这里阅读我以前的故事：

Recently, I come across a medium post on announcement off Official Release of bar_chart_race by Ted Petrou. In his article, he provides an excellent tutorial on how to create Bar Chart Race using bar_chart_race package. Check out the official document here.

最近，我在Ted Petrou 的bar_chart_race官方发布公告上看到了一个中等职位。 在他的文章中，他提供了关于如何使用bar_chart_race包创建条形图竞赛的出色教程。 在此处查看正式文件。

内容 (Content)

• Installation of Bar Chart Race package
  安装条形图竞赛套件
• Import required libraries
  导入所需的库
• Loading Data
  加载数据中
• Creating Bar Chart Race
  创建条形图竞赛

安装条形图竞赛套件 (Installation of Bar Chart Race package)

Install with pip:

用pip安装：

pip3 install bar_chart_race

or using anaconda:

或使用水蟒：

conda install -c conda-forge bar_chart_race

导入所需的库 (Import Required Libraries)

For this project, we only need Pandas, Bar_Chart_Race libraries.

对于此项目，我们仅需要Pandas，Bar_Chart_Race库。

# to load data
import pandas as pd#bar chart race
import bar_chart_race as bcr

加载数据集 (Loading Dataset)

%time population = pd.read_csv('./data/Countries Population from 1995 to 2020.csv')

In bar chart race, your data must be in a specific format:

在条形图竞赛中，您的数据必须采用特定格式：

• each entry represents a single time
  每个条目代表一个时间
• each feature have some single particular value
  每个功能都有一些特定的价值
• time should be set as .index
  时间应设置为.index

Let’s have a look at how our data is looking.

让我们看一下数据的外观。

population.head()

So it’s clear that our data is not in the appropriate format to feed in bar_chart_race. First, make relevant changes in data.

因此很明显，我们的数据格式不适合输入bar_chart_race。 首先，对数据进行相关更改。

Step 1 — Remove all columns except Year, Country, and Population.

步骤1 —删除“年”，“国家”和“人口”以外的所有列。

pop = population.drop(['Yearly % Change', 'Yearly Change', 'Migrants (net)', 'Median Age', 'Fertility Rate', 'Density (P/Km²)', 'Urban Pop %', 'Urban Population', 'Country\'s Share of World Pop %', 'World Population', 'Country Global Rank' ], axis = 1)

Step 2 — Create pivot_table from the pop data frame where Year is an index; each country as column and Population as value.

第2步 -从Year为索引的弹出数据框中创建ivot_table； 每个国家/地区为栏，人口为数值。

df = pop.pivot_table('Population',['Year'], 'Country')

Sometimes your data is not in order, so make sure you order the time column. In our case, its Year.

有时您的数据不整齐，因此请确保您对时间列进行了排序。 在我们的例子中，它是Year 。

df.sort_values(list(df.columns),inplace=True)
df = df.sort_index()
df

Now our data is ready, so let’s create a bar chart race.

现在我们的数据已经准备好了，让我们创建一个条形图竞赛。

创建条形图竞赛 (Create Bar Chart Race)

You can simply use .bar_chart_race()method from bcr.

您可以简单地使用bcr .bar_chart_race()方法。

bcr.bar_chart_race(df)

The above give step is very simple and not as attractive as I want. So let’s customize it. Let’s have a look at the final code. You can find all features and different possibilities in bar_chart_plot documentation.

上面给出的步骤非常简单，没有我想要的那么吸引人。 因此，让我们对其进行自定义。 让我们看一下最终代码。 您可以在bar_chart_plot文档中找到所有功能和不同的可能性。

bcr.bar_chart_race(
    df=df,
    filename=None,
    orientation='h',
    sort='desc',
    n_bars=10,
    fixed_order=False,
    fixed_max=True,
    steps_per_period=10,
    interpolate_period=False,
    label_bars=True,
    bar_size=.90,
    period_label={'x': .99, 'y': .25, 'ha': 'right', 'va':'center'},
    period_summary_func=lambda v, r: {'x': .99, 'y': .18,
             's': f'Population{v.nlargest(39).sum():,.0f}',
             'ha': 'right', 'size': 8, 'family': 'Courier New'},
    period_length=500,
    figsize=(6.5,5),
    dpi=144,
    cmap='dark12',
    title='Population by Country',
    title_size='',
    bar_label_size=7,
    tick_label_size=5,
    shared_fontdict={'family' : 'Helvetica','color' : '.1'},
    scale='linear',
    writer=None,
    fig=None,
    bar_kwargs={'alpha': .7},
    filter_column_colors=True)

I would like to thanks Ted Petrou for creating such a great library and Tutorials. Apart from that, thanks to Vinay Shaw — author of this kernel, who introduced me to this beautiful library.

我要感谢Ted Petrou创建了如此出色的库和教程。 除此之外，还要感谢Vinay Shaw(该内核的作者)，他向我介绍了这个美丽的库。

最后的话 (Final Words)

That’s it for today. Today, I learned and created my first bar chart race visualization. In future looking forward to create something interesting from it. I hope this article was helpful to you. All the codes of data analysis and visuals can be found at this GitHub repository or Kaggle kernel.

今天就这样。 今天，我学习并创建了我的第一个条形图种族可视化。 将来希望从中创建一些有趣的东西。 希望本文对您有所帮助。 可以在此GitHub存储库或Kaggle内核中找到所有数据分析和可视化代码。

Thanks for reading.

谢谢阅读。

I appreciate any feedback.

我感谢任何反馈。

数据科学进展100天 (100 Days of Data Science Progress)

If you like my work and want to support me, I’d greatly appreciate it if you follow me on my social media channels:

如果您喜欢我的工作并希望支持我，那么如果您在我的社交媒体频道上关注我，我将不胜感激：

• The best way to support me is by following me on Medium.

  支持我的最佳方法是在Medium上关注我。

• Subscribe to my new YouTube channel.

  订阅我的新YouTube频道 。

• Sign up on my email list.

  在我的电子邮件列表中注册。