Welcome to The Beginner’s Guide to Building Interactive Maps in Python

欢迎使用Python构建交互式地图的初学者指南

In this post, I would like to show you how to create interactive climate maps using the Historical Climate Data, where you can visualize, examine, and explore the data. Data visualization plays an important role in representing data. Creating visualizations helps to present your analysis in an easier form of understanding. Especially when working with large datasets it is very easy to get lost, that’s when we can see the power of data visualization. In this exercise, we will work with climate data from Kaggle. We will build two interactive climate maps. The first one will be showing the climate change of each country, and the second one will be showing the temperature change over time. Let’s get started, we have a lot to do!

在本文中，我想向您展示如何使用历史气候数据创建交互式气候图，您可以在其中可视化，检查和探索数据。 数据可视化在表示数据中起着重要作用。 创建可视化有助于以一种更容易理解的方式呈现您的分析。 特别是在处理大型数据集时，很容易迷失方向，这就是我们可以看到数据可视化的强大功能。 在本练习中，我们将使用来自Kaggle的气候数据。 我们将构建两个交互式气候图。 第一个显示每个国家的气候变化，第二个显示随着时间的温度变化。 让我们开始吧，我们还有很多事要做！

Kaggle is the world’s largest data science community with powerful tools and resources to help you achieve your data science goals.

Kaggle是全球最大的数据科学社区，其功能强大的工具和资源可帮助您实现数据科学目标。

目录： (Table of Contents:)

• Plotly密谋
• Understanding the Data了解数据
• Data Cleaning数据清理
• Data Preprocessing数据预处理
• Data Visualization数据可视化

密谋 (Plotly)

Plotly is Python graphing library that makes interactive, publication-quality graphs. Examples of how to make line plots, scatter plots, area charts, bar charts, error bars, box plots, histograms, heatmaps, subplots, multiple-axes, polar charts, and bubble charts. It is also an open-source library.

Plotly是Python图形库，可制作交互式的，具有出版质量的图形。 有关如何制作折线图，散点图，面积图，条形图，误差线，箱形图，直方图，热图，子图，多轴图，极坐标图和气泡图的示例。 它也是一个开源库。

To learn more about Plotly: Plotly Graphing Library

要了解有关Plotly的更多信息： Plotly Graphing Library

了解数据 (Understanding the Data)

The Berkeley Earth Surface Temperature Study combines 1.6 billion temperature reports from 16 pre-existing archives. It is nicely packaged and allows for slicing into interesting subsets (for example by country). They publish the source data and the code for the transformations they applied.

伯克利地球表面温度研究结合了16个现有档案中的16亿个温度报告。 它包装精美，可以切成有趣的子集(例如，按国家/地区)。 他们为应用的转换发布源数据和代码。

Dataset can be found at the following link: Climate Data

可以在以下链接中找到数据集： 气候数据

The data folder includes the following datasets:

数据文件夹包含以下数据集：

• Global Average Land Temperature by Country全球平均陆地温度(按国家)
• Global Average Land Temperature by State全球各州平均陆地温度
• Global Land Temperatures By Major City主要城市的全球陆地温度
• Global Land Temperatures By City全球城市气温
• Global Land and Ocean-and-Land Temperatures全球陆地和海洋和陆地温度

We will be working with the “Global Average Land Temperature by Country” dataset, this data fits better for our goal because we are going to build interactive climate maps, and having a data filtered by country will make our life much easier.

我们将使用“按国家/地区划分的全球平均陆地温度”数据集，此数据更适合我们的目标，因为我们将构建交互式气候图，并且按国家/地区过滤数据将使我们的生活变得更加轻松。

图书馆 (Libraries)

We will need three main libraries to get started. When we come to visualization I will ask you to import a couple more sub-libraries, which are also known as library components. For now, we are going to import the following libraries:

我们将需要三个主要的库来开始。 进行可视化时，我将要求您导入几个子库，这些子库也称为库组件。 现在，我们将导入以下库：

import numpy as npimport pandas as pdimport plotly as py

If you don’t have these libraries, don’t worry. It is super easy to install them, as you can see below:

如果您没有这些库，请不要担心。 安装它们非常容易，如下所示：

pip install numpy pandas plotly

读取数据 (Read Data)

df = pd.read_csv("data/GlobalLandTemperaturesByCountry.csv")print(df.head())

print(df.tail())

# Checking the null values in each columndf.isnull().sum()

数据清理 (Data Cleaning)

Data Science is more about understanding the data, and data cleaning is a very important part of this process. What makes the data more valuable depends on how much we can get from it. Preparing the data well will make your data analysis results more accurate.

数据科学更多地是关于理解数据的，数据清理是此过程中非常重要的一部分。 什么使数据更有价值取决于我们可以从中获得多少。 做好数据准备将使您的数据分析结果更加准确。

Let’s start with cleaning process. Firstly, let’s start by dropping the “AverageTemperatureUncertainty” column, because we don’t need it.

让我们从清洁过程开始。 首先，让我们开始删除“ AverageTemperatureUncertainty ”列，因为我们不需要它。

df = df.drop("AverageTemperatureUncertainty", axis=1)

Then, let’s rename the column names to have a better look. As you can see above, we are using a method called rename. Isn’t that cool how easy to rename a column name.

然后，让我们重命名列名称以使其外观更好。 如您在上面所看到的，我们正在使用一种称为重命名的方法。 重命名列名称的难易程度不是很酷。

df = df.rename(columns={'dt':'Date'})df = df.rename(columns={'AverageTemperature':'AvTemp'})

Lastly for the data cleaning, let’s drop the rows with the null values so that they don’t effect our analysis. As we checked earlier, we have around 32000 rows with null values in AverageTemperature column. And in total we have around 577000 rows, so dropping them is not a big deal. But in some cases, there are a couple other methods to handle null values.

最后，为了进行数据清理，让我们删除具有空值的行，以免影响我们的分析。 正如我们之前所检查的，AverageTemperature列中大约有32000行具有空值。 总共有大约577000行，因此删除它们并不是什么大问题。 但是在某些情况下，还有其他几种方法可以处理空值。

df = df.dropna()

Now, let’s have a look at our dataframe. I will print the first 10 rows using the head method.

现在，让我们看一下我们的数据框。 我将使用head方法打印前10行。

df.head(10)

数据预处理 (Data Preprocessing)

This step is also known as data manipulation, where we filter the data so that we can focus on a specific analysis. Especially when working with big datasets, data preprocessing/ filtering is a must. For example, our historical climate data is showing temperatures for all 12 months between 1744 to 2013, so it’s actually a very wide range. Using data filtering techniques, we will focus on a smaller range like between 2000 to 2002.

此步骤也称为数据处理，其中我们对数据进行过滤，以便我们可以专注于特定的分析。 特别是在处理大型数据集时，必须进行数据预处理/过滤。 例如，我们的历史气候数据显示了1744年至2013年之间的所有12个月的温度，因此实际上范围很广。 使用数据过滤技术，我们将专注于较小的范围，例如2000到2002年之间。

比较运算符 (Comparison Operators)

• <<
• >>
• <=<=
• >=> =
• ====
• !=！=

We will use these operators to compare a specific value to values in the column. The result will be a series of booleans: True and Falses. True if the comparison is right, false if the comparison is not right.

我们将使用这些运算符将特定值与列中的值进行比较。 结果将是一系列布尔值：True和Falses。 如果比较正确，则为true；如果比较不正确，则为false。

分组依据 (Grouping by)

In this step, we are grouping the dataframe by Country name and the date columns. And also, sorting the values by date from latest to earliest time.

在此步骤中，我们将按国家/地区名称和日期列对数据框进行分组。 而且，还可以按日期从最晚到最早的时间对值进行排序。

df_countries = df.groupby( ['Country','Date']).sum().reset_index().sort_values('Date', ascending=False)

屏蔽数据范围 (Masking by the data range)

start_date = '2000-01-01' end_date = '2002-01-01' mask = (df_countries['Date'] > start_date) & (df_countries['Date'] <= end_date) df_countries = df_countries.loc[mask] df_countries.head(10)

As you can see above, the dataframe is looking great. Sorted by date and filtered by country name. We can find the average temperature in each month of each country by looking at this dataframe. Here comes the fun part, which is data visualization. Are you ready?

正如您在上面看到的，数据框看起来很棒。 按日期排序并按国家/地区名称过滤。 通过查看此数据框，我们可以找到每个国家/地区每个月的平均温度。 这是有趣的部分，它是数据可视化。 你准备好了吗？

数据可视化 (Data Visualization)

情节的组成 (Components of Plotly)

Before we start, as mentioned earlier there are couple sub-libraries to import to enjoy data visualization. These sub-libraries are also known as Components.

在开始之前，如前所述，有几个子库需要导入才能享受数据可视化。 这些子库也称为组件。

#Plotly Componentsimport plotly.express as pximport plotly.graph_objs as gofrom plotly.subplots import make_subplotsfrom plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot

气候变化图 (Climate Change Map)

Perfect, now by running the following code you will see the magic happening.

完美，现在通过运行以下代码，您将看到魔术的发生。

#Creating the visualizationfig = go.Figure(data=go.Choropleth( locations = df_countries['Country'], locationmode = 'country names', z = df_countries['AvTemp'], colorscale = 'Reds', marker_line_color = 'black', marker_line_width = 0.5, ))fig.update_layout( title_text = 'Climate Change', title_x = 0.5, geo=dict( showframe = False, showcoastlines = False, projection_type = 'equirectangular' ) ) fig.show()

气候变化的时间表 (Climate Change by Timeline)

# Manipulating the original dataframe df_countrydate = df_countries.groupby(['Date','Country']). sum().reset_index() #Creating the visualization fig = px.choropleth(df_countrydate, locations="Country", locationmode = "country names", color="AvTemp", hover_name="Country", animation_frame="Date" ) fig.update_layout( title_text = 'Average Temperature Change', title_x = 0.5, geo=dict( showframe = False, showcoastlines = False, )) fig.show()

结果 (Results)

Both are the same map, in the first one you can see the change in average temperature. And in the second graph, I am just hovering over some countries, which is showing more detailed information about each of them.

两者是同一张图，在第一个图中，您可以看到平均温度的变化。 在第二张图中，我只是将鼠标悬停在某些国家/地区上，该国家/地区显示了有关每个国家/地区的更详细的信息。

Thank you for reading this post, I hope you enjoyed and learn something new today. Feel free to contact me through my blog if you have any questions while implementing the code. I will be more than happy to help. You can find more posts I’ve published related to Python and Machine Learning. Stay safe and happy coding!

感谢您阅读这篇文章，希望您今天喜欢并学到一些新东西。 如果在实施代码时有任何疑问，请随时通过我的博客与我联系 。 我将非常乐意提供帮助。 您可以找到我发布的更多有关Python和机器学习的文章。 保持安全快乐的编码！

I am Behic Guven, and I love sharing stories on creativity, programming, motivation, and life.

我是Behic Guven，我喜欢分享有关创造力，编程，动力和生活的故事。

Follow my blog and Towards Data Science to stay inspired.

关注 我的博客 和 迈向数据科学 ，保持灵感。

相关文章 (Related Posts)