用python进行营销分析_用python进行covid 19分析
用python進(jìn)行營銷分析
Python is a highly powerful general purpose programming language which can be easily learned and provides data scientists a wide variety of tools and packages. Amid this pandemic period, I decided to do an analysis on this novel coronavirus.
Python是一種功能強(qiáng)大的通用編程語言，可以輕松學(xué)習(xí)，并為數(shù)據(jù)科學(xué)家提供各種工具和軟件包。 在這個(gè)大流行時(shí)期，我決定對這種新型冠狀病毒進(jìn)行分析。
In this article, I am going to walk you through the steps I undertook for this analysis with visuals and code snippets.
在本文中，我將通過視覺和代碼片段逐步指導(dǎo)您進(jìn)行此分析。
數(shù)據(jù)分析涉及的步驟： (Steps involved in Data Analysis:)
2. Gathering Data
2.收集數(shù)據(jù)
3. Transforming Data to our needs (Data Wrangling)
3.將數(shù)據(jù)轉(zhuǎn)變?yōu)槲覀兊男枨?數(shù)據(jù)整理)
4. Exploratory Data Analysis (EDA) and Visualization
4.探索性數(shù)據(jù)分析(EDA)和可視化
步驟— 1：導(dǎo)入所需的軟件包 (Step — 1: Importing required Packages)
Importing our required packages is the starting point of all data analysis programming in python. As I’ve said, python provides a wide variety of packages for data scientists and in this analysis I used python’s most popular data science packages Pandas and NumPy for Data Wrangling and EDA. When coming to Data Visualization, I used python’s interactive packages Plotly and Matplotlib. It’s very simple to import packages in python code:
導(dǎo)入所需的軟件包是python中所有數(shù)據(jù)分析編程的起點(diǎn)。 就像我說過的那樣，python為數(shù)據(jù)科學(xué)家提供了各種各樣的軟件包，在此分析中，我使用了python最受歡迎的數(shù)據(jù)科學(xué)軟件包Pandas和NumPy進(jìn)行數(shù)據(jù)整理和EDA。 進(jìn)行數(shù)據(jù)可視化時(shí)，我使用了python的交互式軟件包Plotly和Matplotlib。 用python代碼導(dǎo)入軟件包非常簡單：
This is the code for importing our primary packages to perform data analysis but still, we need to add some more packages to our code which we will see in step-2. Yay! We successfully finished our first step.
這是用于導(dǎo)入主要程序包以執(zhí)行數(shù)據(jù)分析的代碼，但是仍然需要向代碼中添加更多程序包，我們將在步驟2中看到這些代碼。 好極了！ 我們成功地完成了第一步。
步驟2：收集數(shù)據(jù) (Step — 2: Gathering Data)
For a clean and perfect data analysis, the foremost important element is collecting quality Data. For this analysis, I’ve collected many data from various sources for better accuracy.
對于 干凈，完美的數(shù)據(jù)分析，最重要的元素是收集高質(zhì)量的數(shù)據(jù)。 為了進(jìn)行此分析，我從各種來源收集了許多數(shù)據(jù)，以提高準(zhǔn)確性。
Our primary dataset is extracted from esri (a website which provides updated data on coronavirus) using a query url (click here to view the website). Follow the code snippets to extract the data from esri:
我們的主要數(shù)據(jù)集是使用查詢網(wǎng)址從esri(提供有關(guān)冠狀病毒的最新數(shù)據(jù)的網(wǎng)站)中提取的( 請單擊此處查看該網(wǎng)站 )。 按照代碼片段從esri中提取數(shù)據(jù)：
Requests is a python packages used to extract data from a given json file. In this code I used requests to extract data from the given query url by esri. We are now ready to do some Data Wrangling! (Note : We will be importing many data in step-4 of our analysis)
Requests是一個(gè)python軟件包，用于從給定的json文件中提取數(shù)據(jù)。 在這段代碼中，我使用了esri的請求從給定的查詢URL中提取數(shù)據(jù)。 現(xiàn)在，我們準(zhǔn)備進(jìn)行一些數(shù)據(jù)整理！ (注意：我們將在分析的第4步中導(dǎo)入許多數(shù)據(jù))
步驟— 3：數(shù)據(jù)整理 (Step — 3: Data Wrangling)
Data Wrangling is a process where we will transform and clean our data to our needs. We can’t do analysis with our raw extracted data. So, we have to transform the data to proceed our analysis. Here’s the code for our Data Wrangling:
數(shù)據(jù)整理是一個(gè)過程，在此過程中，我們將根據(jù)需要轉(zhuǎn)換和清理數(shù)據(jù)。 我們無法使用原始提取的數(shù)據(jù)進(jìn)行分析。 因此，我們必須轉(zhuǎn)換數(shù)據(jù)以進(jìn)行分析。 這是我們的數(shù)據(jù)整理的代碼：
Note that, we have imported a new python package, ‘datetime’, which helps us to work with dates and times in a dataset. Now, get ready to see the big picture of our analysis -’ EDA and Data Visualization’.
請注意，我們已經(jīng)導(dǎo)入了一個(gè)新的python包“ datetime”，它可以幫助我們處理數(shù)據(jù)集中的日期和時(shí)間。 現(xiàn)在，準(zhǔn)備看一下我們分析的大圖-“ EDA和數(shù)據(jù)可視化”。
步驟— 4：探索性數(shù)據(jù)分析和數(shù)據(jù)可視化 (Step — 4: Exploratory Data Analysis and Data Visualization)
This process is quite long as it is the heart and soul of data analysis. So, I’ve divided this process into three steps:
這個(gè)過程很長，因?yàn)樗菙?shù)據(jù)分析的心臟和靈魂。 因此，我將這一過程分為三個(gè)步驟：
a. Ranking countries and provinces (based on COVID-19 aspects)
一個(gè)。 對國家和省進(jìn)行排名(基于COVID-19方面)
b. Time Series on COVID-19 Cases
b。 COVID-19病例的時(shí)間序列
c. Classification and Distribution of cases
C。 案件分類和分布
Ranking countries and provinces
排名國家和省
From our previously extracted data we are going to rank countries and provinces based on confirmed, deaths, recovered and active cases by doing some EDA and Visualization. Follow the code snippets for the upcoming visuals (Note : Every visualizations are interactive and you can hover them to see their data points)
從我們先前提取的數(shù)據(jù)中，我們將通過進(jìn)行一些EDA和可視化，根據(jù)確診，死亡，康復(fù)和活著的病例對國家和省進(jìn)行排名。 請遵循即將出現(xiàn)的視覺效果的代碼片段(注意：每個(gè)視覺效果都是交互式的，您可以將它們懸停以查看其數(shù)據(jù)點(diǎn))
Part 1 — Ranking Most affected countries
第1部分-排名受影響最大的國家
i) Top 10 Confirmed Cases Countries:
i)十大確診病例國家：
The following code will produce a plot ranking top 10 countries based on confirmed cases.
以下代碼將根據(jù)已確認(rèn)的案例得出前十個(gè)國家/地區(qū)的地塊。
# a. Top 10 confirmed countries (Bubble plot)top10_confirmed = pd.DataFrame(data.groupby('Country')['Confirmed'].sum().nlargest(10).sort_values(ascending = False)) fig1 = px.scatter(top10_confirmed, x = top10_confirmed.index, y = 'Confirmed', size = 'Confirmed', size_max = 120,color = top10_confirmed.index, title = 'Top 10 Confirmed Cases Countries') fig1.show()演示地址
Graph by Author作者圖ii) Top 10 Death Cases Countries:
ii)十大死亡案例國家：
The following code will produce a plot ranking top 10 countries based on death cases.
以下代碼將根據(jù)死亡案例產(chǎn)生一個(gè)排在前十個(gè)國家/地區(qū)的地塊。
# b. Top 10 deaths countries (h-Bar plot)top10_deaths = pd.DataFrame(data.groupby('Country')['Deaths'].sum().nlargest(10).sort_values(ascending = True)) fig2 = px.bar(top10_deaths, x = 'Deaths', y = top10_deaths.index, height = 600, color = 'Deaths', orientation = 'h',color_continuous_scale = ['deepskyblue','red'], title = 'Top 10 Death Cases Countries') fig2.show()演示地址
Graph by Author作者圖iii) Top 10 Recovered Cases Countries:
iii)十大被追回病例國家：
The following code will produce a plot ranking top 10 countries based on recovered cases.
以下代碼將根據(jù)恢復(fù)的案例生成一個(gè)排在前10個(gè)國家/地區(qū)的地塊。
# c. Top 10 recovered countries (Bar plot)top10_recovered = pd.DataFrame(data.groupby('Country')['Recovered'].sum().nlargest(10).sort_values(ascending = False)) fig3 = px.bar(top10_recovered, x = top10_recovered.index, y = 'Recovered', height = 600, color = 'Recovered',title = 'Top 10 Recovered Cases Countries', color_continuous_scale = px.colors.sequential.Viridis) fig3.show()演示地址
Graph by Author作者圖iv) Top 10 Active Cases Countries:
iv)十大活躍案例國家：
The following code will produce a plot ranking top 10 countries based on recovered cases.
以下代碼將根據(jù)恢復(fù)的案例生成一個(gè)排在前10個(gè)國家/地區(qū)的地塊。
# d. Top 10 active countriestop10_active = pd.DataFrame(data.groupby('Country')['Active'].sum().nlargest(10).sort_values(ascending = True)) fig4 = px.bar(top10_active, x = 'Active', y = top10_active.index, height = 600, color = 'Active', orientation = 'h',color_continuous_scale = ['paleturquoise','blue'], title = 'Top 10 Active Cases Countries') fig4.show()演示地址
Graph by Author作者圖Part 2 — Ranking most affected States in largely affected Countries:
第2部分-在受影響最大的國家中對受影響最大的國家進(jìn)行排名：
EDA for ranking states in largely affected Countries:
對受災(zāi)嚴(yán)重國家排名的EDA：
# USA topstates_us = data['Country'] == 'US' topstates_us = data[topstates_us].nlargest(5, 'Confirmed') # Brazil topstates_brazil = data['Country'] == 'Brazil' topstates_brazil = data[topstates_brazil].nlargest(5, 'Confirmed') # India topstates_india = data['Country'] == 'India' topstates_india = data[topstates_india].nlargest(5, 'Confirmed') # Russia topstates_russia = data['Country'] == 'Russia' topstates_russia = data[topstates_russia].nlargest(5, 'Confirmed')We are extracting States’ data from USA, Brazil, India and Russia respectively because, these are the countries which are most affected by COVID-19. Now, let’s visualize it!
我們分別從美國，巴西，印度和俄羅斯提取州的數(shù)據(jù)，因?yàn)檫@是受COVID-19影響最大的國家。 現(xiàn)在，讓我們對其進(jìn)行可視化！
Visualization of Most affected states in largely affected Countries:
受影響最大的國家中受影響最大的州的可視化：
i) Most affected States in USA:
i)美國受影響最嚴(yán)重的州：
The following code will produce a plot ranking top 5 most affected states in the United States of America.
以下代碼將產(chǎn)生一個(gè)在美國受災(zāi)最嚴(yán)重的州中排名前5位的地塊。
# USA fig5 = go.Figure(data = [go.Bar(name = 'Active Cases', x = topstates_us['Active'], y = topstates_us['State'], orientation = 'h'),go.Bar(name = 'Death Cases', x = topstates_us['Deaths'], y = topstates_us['State'], orientation = 'h') ]) fig5.update_layout(title = 'Most Affected States in USA', height = 600) fig5.show()演示地址
Graph by Author作者圖ii) Most affected States in Brazil:
ii)巴西受影響最嚴(yán)重的國家：
The following code will produce a plot ranking top 5 most affected states in Brazil.
以下代碼將產(chǎn)生一個(gè)在巴西受影響最嚴(yán)重的州中排名前5位的地塊。
# Brazil fig6 = go.Figure(data = [go.Bar(name = 'Recovered Cases', x = topstates_brazil['State'], y = topstates_brazil['Recovered']),go.Bar(name = 'Active Cases', x = topstates_brazil['State'], y = topstates_brazil['Active']),go.Bar(name = 'Death Cases', x = topstates_brazil['State'], y = topstates_brazil['Deaths']) ]) fig6.update_layout(title = 'Most Affected States in Brazil', barmode = 'stack', height = 600) fig6.show()演示地址
Graph by Author作者圖iii) Most affected States in India:
iii)印度受影響最大的國家：
The following code will produce a plot ranking top 5 most affected states in India.
以下代碼將產(chǎn)生一個(gè)在印度受影響最嚴(yán)重的州中排名前5位的地塊。
# India fig7 = go.Figure(data = [go.Bar(name = 'Recovered Cases', x = topstates_india['State'], y = topstates_india['Recovered']),go.Bar(name = 'Active Cases', x = topstates_india['State'], y = topstates_india['Active']),go.Bar(name = 'Death Cases', x = topstates_india['State'], y = topstates_india['Deaths']) ]) fig7.update_layout(title = 'Most Affected States in India', barmode = 'stack', height = 600) fig7.show()演示地址
Graph by Author作者圖iv) Most affected States in Russia:
iv)俄羅斯受影響最嚴(yán)重的國家：
The following code will produce a plot ranking top 5 most affected states in Russia.
以下代碼將產(chǎn)生一個(gè)在俄羅斯受影響最嚴(yán)重的州中排名前5位的地塊。
# Russia fig8 = go.Figure(data = [go.Bar(name = 'Recovered Cases', x = topstates_russia['State'], y = topstates_russia['Recovered']),go.Bar(name = 'Active Cases', x = topstates_russia['State'], y = topstates_russia['Active']),go.Bar(name = 'Death Cases', x = topstates_russia['State'], y = topstates_russia['Deaths']) ]) fig8.update_layout(title = 'Most Affected States in Russia', barmode = 'stack', height = 600) fig8.show()演示地址
Graph by Author作者圖Time Series on COVID-19 Cases
COVID-19病例的時(shí)間序列
To perform time series analysis on COVID-19 cases we need a new dataset. https://covid19.who.int/ Follow this link and images shown below for downloading our next dataset.
要對COVID-19案例執(zhí)行時(shí)間序列分析，我們需要一個(gè)新的數(shù)據(jù)集。 https://covid19.who.int/請點(diǎn)擊下面的鏈接和圖像，下載我們的下一個(gè)數(shù)據(jù)集。
After pressing the link mentioned above, you will land into this page. On the bottom right of the represented map, you can find the download button. From there you can download the dataset and save it to your files. Good work! We fetched our Data! Let’s import the data :
按下上述鏈接后，您將進(jìn)入此頁面。 在所顯示地圖的右下角，您可以找到下載按鈕。 從那里您可以下載數(shù)據(jù)集并將其保存到文件中。 干得好！ 我們獲取了數(shù)據(jù)！ 讓我們導(dǎo)入數(shù)據(jù)：
time_series = pd.read_csv('who_data.csv', encoding = 'ISO-8859-1') time_series['Date_reported'] = pd.to_datetime(time_series['Date_reported'])From the above extracted dataset, we are going to perform two types of time series analysis, ‘COVID-19 cases Worldwide’ and ‘Most affected countries over time’.
從上面提取的數(shù)據(jù)集中，我們將執(zhí)行兩種類型的時(shí)間序列分析：“全球COVID-19病例”和“一段時(shí)間內(nèi)受影響最大的國家”。
i) COVID-19 cases worldwide:
i)全球COVID-19病例：
EDA for COVID-19 cases worldwide:
全球COVID-19案件的EDA：
time_series_dates = time_series.groupby('Date_reported').sum() time_series_datesa) Cumulative cases worldwide:
a)全世界的累積病例：
The following code produces a time series chart of cumulative cases worldwide right from the beginning of the outbreak.
以下代碼從爆發(fā)開始就生成了全球累積病例的時(shí)序圖。
# Cumulative casesfig11 = go.Figure() fig11.add_trace(go.Scatter(x = time_series_dates.index, y = time_series_dates[' Cumulative_cases'], fill = 'tonexty',line_color = 'blue')) fig11.update_layout(title = 'Cumulative Cases Worldwide') fig11.show()演示地址
Graph by Author作者圖b) Cumulative death cases worldwide:
b)世界范圍內(nèi)的累積死亡案例：
The following code produces a time series chart of cumulative death cases worldwide right from the beginning of the outbreak.
以下代碼從爆發(fā)開始就生成了全球累積死亡病例的時(shí)間序列圖。
# Cumulative death casesfig12 = go.Figure() fig12.add_trace(go.Scatter(x = time_series_dates.index, y = time_series_dates[' Cumulative_deaths'], fill = 'tonexty',line_color = 'red')) fig12.update_layout(title = 'Cumulative Deaths Worldwide') fig12.show()演示地址
Graph by Author作者圖c) Daily new cases worldwide:
c)全世界每天的新病例：
The following code produces a time series chart of daily new cases worldwide right from the beginning of the outbreak.
以下代碼從爆發(fā)開始就生成了全球每日新病例的時(shí)序圖。
# Daily new casesfig13 = go.Figure() fig13.add_trace(go.Scatter(x = time_series_dates.index, y = time_series_dates[' New_cases'], fill = 'tonexty',line_color = 'gold')) fig13.update_layout(title = 'Daily New Cases Worldwide') fig13.show()演示地址
Graph by Author作者圖d) Daily death cases worldwide:
d)全世界每日死亡案例：
The following code produces a time series chart of daily death cases worldwide right from the beginning of the outbreak.
以下代碼從爆發(fā)開始就生成了全球每日死亡病例的時(shí)間序列圖。
# Daily death casesfig14 = go.Figure() fig14.add_trace(go.Scatter(x = time_series_dates.index, y = time_series_dates[' New_deaths'], fill = 'tonexty',line_color = 'hotpink')) fig14.update_layout(title = 'Daily Death Cases Worldwide') fig14.show()演示地址
Graph by Author作者圖ii) Most affected countries over time:
ii)一段時(shí)間內(nèi)受影響最大的國家：
EDA for Most affected countries over time:
隨時(shí)間推移對受影響最嚴(yán)重國家的EDA：
# USA time_series_us = time_series[' Country'] == ('United States of America') time_series_us = time_series[time_series_us]# Brazil time_series_brazil = time_series[' Country'] == ('Brazil') time_series_brazil = time_series[time_series_brazil]# India time_series_india = time_series[' Country'] == ('India') time_series_india = time_series[time_series_india]# Russia time_series_russia = time_series[' Country'] == ('Russia') time_series_russia = time_series[time_series_russia]# Peru time_series_peru = time_series[' Country'] == ('Peru') time_series_peru = time_series[time_series_peru]Note that, we have extracted data of countries USA, Brazil, India, Russia and Peru respectively as they are highly affected by COVID-19 in the world.
請注意，我們分別提取了美國，巴西，印度，俄羅斯和秘魯?shù)葒?地區(qū)的數(shù)據(jù)，因?yàn)樗鼈冊谑澜绶秶鷥?nèi)受到COVID-19的高度影響。
a) Most affected Countries’ Cumulative cases over time
a)隨時(shí)間推移受影響最嚴(yán)重的國家的累計(jì)案件
The following code will produce a time series chart of the most affected countries’ cumulative cases right from the beginning of the outbreak.
以下代碼將從疫情爆發(fā)之初就產(chǎn)生受影響最嚴(yán)重國家累計(jì)病例的時(shí)序圖。
# Cumulative casesfig15 = go.Figure()fig15.add_trace(go.Line(x = time_series_us['Date_reported'], y = time_series_us[' Cumulative_cases'], name = 'USA')) fig15.add_trace(go.Line(x = time_series_brazil['Date_reported'], y = time_series_brazil[' Cumulative_cases'], name = 'Brazil')) fig15.add_trace(go.Line(x = time_series_india['Date_reported'], y = time_series_india[' Cumulative_cases'], name = 'India')) fig15.add_trace(go.Line(x = time_series_russia['Date_reported'], y = time_series_russia[' Cumulative_cases'], name = 'Russia')) fig15.add_trace(go.Line(x = time_series_peru['Date_reported'], y = time_series_peru[' Cumulative_cases'], name = 'Peru'))fig15.update_layout(title = 'Time Series of Most Affected countries"s Cumulative Cases')fig15.show()演示地址
Graph by Author作者圖b) Most affected Countries’ cumulative death cases over time:
b)隨著時(shí)間的推移，大多數(shù)受影響國家的累計(jì)死亡病例：
The following code will produce a time series chart of the most affected countries’ cumulative death cases right from the beginning of the outbreak.
以下代碼將從疫情爆發(fā)之初就繪制出受影響最嚴(yán)重國家累計(jì)死亡病例的時(shí)間序列圖。
# Cumulative death casesfig16 = go.Figure()fig16.add_trace(go.Line(x = time_series_us['Date_reported'], y = time_series_us[' Cumulative_deaths'], name = 'USA')) fig16.add_trace(go.Line(x = time_series_brazil['Date_reported'], y = time_series_brazil[' Cumulative_deaths'], name = 'Brazil')) fig16.add_trace(go.Line(x = time_series_india['Date_reported'], y = time_series_india[' Cumulative_deaths'], name = 'India')) fig16.add_trace(go.Line(x = time_series_russia['Date_reported'], y = time_series_russia[' Cumulative_deaths'], name = 'Russia')) fig16.add_trace(go.Line(x = time_series_peru['Date_reported'], y = time_series_peru[' Cumulative_deaths'], name = 'Peru'))fig16.update_layout(title = 'Time Series of Most Affected countries"s Cumulative Death Cases')fig16.show()演示地址
Graph by Author作者圖c) Most affected Countries’ daily new cases over time:
c)一段時(shí)間以來受影響最嚴(yán)重的國家的每日新病例：
The following code will produce a time series chart of the most affected countries’ daily new cases right from the beginning of the outbreak.
以下代碼將從疫情爆發(fā)之初就產(chǎn)生受影響最嚴(yán)重國家的每日新病例的時(shí)序圖。
# Daily new casesfig17 = go.Figure()fig17.add_trace(go.Line(x = time_series_us['Date_reported'], y = time_series_us[' New_cases'], name = 'USA')) fig17.add_trace(go.Line(x = time_series_brazil['Date_reported'], y = time_series_brazil[' New_cases'], name = 'Brazil')) fig17.add_trace(go.Line(x = time_series_india['Date_reported'], y = time_series_india[' New_cases'], name = 'India')) fig17.add_trace(go.Line(x = time_series_russia['Date_reported'], y = time_series_russia[' New_cases'], name = 'Russia')) fig17.add_trace(go.Line(x = time_series_peru['Date_reported'], y = time_series_peru[' New_cases'], name = 'Peru'))fig17.update_layout(title = 'Time Series of Most Affected countries"s Daily New Cases')fig17.show()演示地址
Graph by Author作者圖d) Most affected Countries’ daily death cases:
d)最受影響國家的每日死亡案例：
The following code will produce a time series chart of the most affected countries’ daily death cases right from the beginning of the outbreak.
以下代碼將從疫情爆發(fā)之初就產(chǎn)生受影響最嚴(yán)重國家的每日死亡病例的時(shí)序圖。
# Daily death casesfig18 = go.Figure()fig18.add_trace(go.Line(x = time_series_us['Date_reported'], y = time_series_us[' New_deaths'], name = 'USA')) fig18.add_trace(go.Line(x = time_series_brazil['Date_reported'], y = time_series_brazil[' New_deaths'], name = 'Brazil')) fig18.add_trace(go.Line(x = time_series_india['Date_reported'], y = time_series_india[' New_deaths'], name = 'India')) fig18.add_trace(go.Line(x = time_series_russia['Date_reported'], y = time_series_russia[' New_deaths'], name = 'Russia')) fig18.add_trace(go.Line(x = time_series_peru['Date_reported'], y = time_series_peru[' New_deaths'], name = 'Peru'))fig18.update_layout(title = 'Time Series of Most Affected countries"s Daily Death Cases')fig18.show()演示地址
Graph by Author作者圖Case Classification and Distribution
病例分類與分布
Here we are going to analyze how COVID-19 cases are distributed. For this, we need a new dataset. https://www.kaggle.com/imdevskp/corona-virus-report Follow this link for our new dataset.
在這里，我們將分析COVID-19案例的分布方式。 為此，我們需要一個(gè)新的數(shù)據(jù)集。 https://www.kaggle.com/imdevskp/corona-virus-report請點(diǎn)擊此鏈接以獲取新數(shù)據(jù)集。
i) WHO Region-Wise Case Distribution:
i)世衛(wèi)組織區(qū)域明智病例分布：
For this analysis, we are going to use ‘country_wise_latest.csv’ dataset which will come along with the downloaded kaggle dataset. The following code produces a pie chart representing case distribution among WHO Region classification.
對于此分析，我們將使用“ country_wise_latest.csv”數(shù)據(jù)集，該數(shù)據(jù)集將與下載的kaggle數(shù)據(jù)集一起提供。 以下代碼生成了一個(gè)餅圖，代表了WHO區(qū)域分類之間的病例分布。
who = pd.read_csv('country_wise_latest.csv') who_region = pd.DataFrame(who.groupby('WHO Region')['Confirmed'].sum())labels = who_region.index values = who_region['Confirmed']fig9 = go.Figure(data=[go.Pie(labels = labels, values = values, pull=[0, 0, 0, 0, 0.2, 0])])fig9.update_layout(title = 'WHO Region-Wise Case Distribution', width = 700, height = 400, margin = dict(t = 0, l = 0, r = 0, b = 0))fig9.show()演示地址
Graph by Author作者圖ii) Most affected Countries’ case distribution:
ii)最受影響國家的案件分布：
For this analysis we are going to use the same ‘country_wise_latest.csv’ dataset which we imported for the previous analysis.
對于此分析，我們將使用為先前的分析導(dǎo)入的相同“ country_wise_latest.csv”數(shù)據(jù)集。
EDA for Most affected countries’ case distribution:
受影響最嚴(yán)重國家的EDA分布：
case_dist = who# US dist_us = case_dist['Country/Region'] == 'US' dist_us = case_dist[dist_us][['Country/Region','Deaths','Recovered','Active']].set_index('Country/Region')# Brazil dist_brazil = case_dist['Country/Region'] == 'Brazil' dist_brazil = case_dist[dist_brazil][['Country/Region','Deaths','Recovered','Active']].set_index('Country/Region')# India dist_india = case_dist['Country/Region'] == 'India' dist_india = case_dist[dist_india][['Country/Region','Deaths','Recovered','Active']].set_index('Country/Region')# Russia dist_russia = case_dist['Country/Region'] == 'Russia' dist_russia = case_dist[dist_russia][['Country/Region','Deaths','Recovered','Active']].set_index('Country/Region')The following code will produce a pie chart representing the case classification on Most affected Countries.
以下代碼將生成一個(gè)餅狀圖，代表大多數(shù)受影響國家/地區(qū)的案件分類。
fig = plt.figure(figsize = (22,14)) colors_series = ['deepskyblue','gold','springgreen','coral'] explode = (0,0,0.1)plt.subplot(221) plt.pie(dist_us, labels = dist_us.columns, colors = colors_series, explode = explode,startangle = 90,autopct = '%.1f%%', shadow = True) plt.title('USA', fontsize = 16)plt.subplot(222) plt.pie(dist_brazil, labels = dist_brazil.columns, colors = colors_series, explode = explode,startangle = 90,autopct = '%.1f%%',shadow = True) plt.title('Brazil', fontsize = 16)plt.subplot(223) plt.pie(dist_india, labels = dist_india.columns, colors = colors_series, explode = explode, startangle = 90, autopct = '%.1f%%',shadow = True) plt.title('India', fontsize = 16)plt.subplot(224) plt.pie(dist_russia, labels = dist_russia.columns, colors = colors_series, explode = explode, startangle = 90,autopct = '%.1f%%', shadow = True) plt.title('Russia', fontsize = 16)plt.suptitle('Case Classification of Most Affected Countries', fontsize = 20)Graph by Author作者圖iii) Most affected continents’ Negative case vs Positive case percentage composition:
iii)受災(zāi)最嚴(yán)重的大陸的消極案例與積極案例的百分比構(gòu)成：
For this analysis we need a new dataset. https://ourworldindata.org/coronavirus-source-data Follow this link to get our next dataset.
對于此分析，我們需要一個(gè)新的數(shù)據(jù)集。 https://ourworldindata.org/coronavirus-source-data單擊此鏈接以獲取我們的下一個(gè)數(shù)據(jù)集。
EDA for Negative case vs Positive case percentage composition :
負(fù)面案例與正面案例所占百分比的EDA：
negative_positive = pd.read_csv('owid-covid-data.csv') negative_positive = negative_positive.groupby('continent')[['total_cases','total_tests']].sum()explode = (0,0.1) labels = ['Postive','Negative'] colors = ['beige','paleturquoise']The following code will produce a pie chart illustrating the percentage composition of Negative cases and Positive cases in most affected Continents.
以下代碼將產(chǎn)生一個(gè)餅圖，說明在受影響最大的大陸中，陰性案例和陽性案例的百分比構(gòu)成。
fig = plt.figure(figsize = (20,20))plt.subplot(321) plt.pie(negative_positive[negative_positive.index == 'Asia'],labels = labels, explode = explode, autopct = '%.1f%%', startangle = 90, colors = colors, shadow = True) plt.title('Asia', fontsize = 15)plt.subplot(322) plt.pie(negative_positive[negative_positive.index == 'North America'],labels = labels, explode = explode, autopct = '%.1f%%', startangle = 90, colors = colors, shadow = True) plt.title('North America', fontsize = 15)plt.subplot(323) plt.pie(negative_positive[negative_positive.index == 'South America'],labels = labels, explode = explode, autopct = '%.1f%%', startangle = 90, colors = colors, shadow = True) plt.title('South America', fontsize = 15)plt.subplot(324) plt.pie(negative_positive[negative_positive.index == 'Europe'],labels = labels, explode = explode, autopct = '%.1f%%', startangle = 90, colors = colors, shadow = True) plt.title('Europe', fontsize = 15)plt.suptitle('Continent-Wise Tested Positive & Negative Percentage Composition', fontsize = 20)Graph by Author作者圖結(jié)論 (Conclusion)
Hurrah! We successfully completed creating our own COVID-19 report with Python. If you forgot to follow any above mentioned steps I have provided the full code for this analysis below. Apart from our analysis, there are much more you can do with Python and its powerful packages. So don’t stop exploring and create your own reports and dashboards.
歡呼！ 我們已經(jīng)成功地使用Python創(chuàng)建了自己的COVID-19報(bào)告。 如果您忘記了執(zhí)行上述任何步驟，則在下面提供了此分析的完整代碼。 除了我們的分析之外，Python及其強(qiáng)大的軟件包還可以做更多的事情。 因此，不要停止探索并創(chuàng)建自己的報(bào)告和儀表板。
You can find many useful resources on internet based on data science in python for example edX, Coursera, Udemy and so on but, never ever stop learning. Hope you find this article useful and knowledgeable. Happy Analyzing!
您可以在python中基于數(shù)據(jù)科學(xué)在Internet上找到許多有用的資源，例如edX，Coursera，Udemy等，但永遠(yuǎn)都不要停止學(xué)習(xí)。 希望您發(fā)現(xiàn)本文有用且知識(shí)淵博。 分析愉快！
FULL CODE:
完整代碼：
import pandas as pd import matplotlib.pyplot as plt import plotly.express as px import numpy as np import plotly import plotly.graph_objects as go import datetime as dt import requests from plotly.subplots import make_subplots# Getting Dataurl_request = requests.get("https://services1.arcgis.com/0MSEUqKaxRlEPj5g/arcgis/rest/services/Coronavirus_2019_nCoV_Cases/FeatureServer/1/query?where=1%3D1&outFields=*&outSR=4326&f=json") url_json = url_request.json() df = pd.DataFrame(url_json['features']) df['attributes'][0]# Data Wrangling# a. transforming datadata_list = df['attributes'].tolist() data = pd.DataFrame(data_list) data.set_index('OBJECTID') data = data[['Province_State','Country_Region','Last_Update','Lat','Long_','Confirmed','Recovered','Deaths','Active']] data.columns = ('State','Country','Last Update','Lat','Long','Confirmed','Recovered','Deaths','Active') data['State'].fillna(value = '', inplace = True) data# b. cleaning datadef convert_time(t):t = int(t)return dt.datetime.fromtimestamp(t)data = data.dropna(subset = ['Last Update']) data['Last Update'] = data['Last Update']/1000 data['Last Update'] = data['Last Update'].apply(convert_time) data# Exploratory Data Analysis & Visualization# Our analysis contains 'Ranking countries and provinces', 'Time Series' and 'Classification and Distribution'# 1. Ranking countries and provinces # a. Top 10 confirmed countries (Bubble plot)top10_confirmed = pd.DataFrame(data.groupby('Country')['Confirmed'].sum().nlargest(10).sort_values(ascending = False)) fig1 = px.scatter(top10_confirmed, x = top10_confirmed.index, y = 'Confirmed', size = 'Confirmed', size_max = 120,color = top10_confirmed.index, title = 'Top 10 Confirmed Cases Countries') fig1.show()# b. Top 10 deaths countries (h-Bar plot)top10_deaths = pd.DataFrame(data.groupby('Country')['Deaths'].sum().nlargest(10).sort_values(ascending = True)) fig2 = px.bar(top10_deaths, x = 'Deaths', y = top10_deaths.index, height = 600, color = 'Deaths', orientation = 'h',color_continuous_scale = ['deepskyblue','red'], title = 'Top 10 Death Cases Countries') fig2.show()# c. Top 10 recovered countries (Bar plot)top10_recovered = pd.DataFrame(data.groupby('Country')['Recovered'].sum().nlargest(10).sort_values(ascending = False)) fig3 = px.bar(top10_recovered, x = top10_recovered.index, y = 'Recovered', height = 600, color = 'Recovered',title = 'Top 10 Recovered Cases Countries', color_continuous_scale = px.colors.sequential.Viridis) fig3.show()# d. Top 10 active countriestop10_active = pd.DataFrame(data.groupby('Country')['Active'].sum().nlargest(10).sort_values(ascending = True)) fig4 = px.bar(top10_active, x = 'Active', y = top10_active.index, height = 600, color = 'Active', orientation = 'h',color_continuous_scale = ['paleturquoise','blue'], title = 'Top 10 Active Cases Countries') fig4.show()# e. Most affected states/provinces in largely affected countries # Here we are going to extract top 4 affected countries' states data and plot it!# Firstly, aggregating data with our dataset : # USA topstates_us = data['Country'] == 'US' topstates_us = data[topstates_us].nlargest(5, 'Confirmed') # Brazil topstates_brazil = data['Country'] == 'Brazil' topstates_brazil = data[topstates_brazil].nlargest(5, 'Confirmed') # India topstates_india = data['Country'] == 'India' topstates_india = data[topstates_india].nlargest(5, 'Confirmed') # Russia topstates_russia = data['Country'] == 'Russia' topstates_russia = data[topstates_russia].nlargest(5, 'Confirmed')# Let's plot! # USA fig5 = go.Figure(data = [go.Bar(name = 'Active Cases', x = topstates_us['Active'], y = topstates_us['State'], orientation = 'h'),go.Bar(name = 'Death Cases', x = topstates_us['Deaths'], y = topstates_us['State'], orientation = 'h') ]) fig5.update_layout(title = 'Most Affected States in USA', height = 600) fig5.show() # Brazil fig6 = go.Figure(data = [go.Bar(name = 'Recovered Cases', x = topstates_brazil['State'], y = topstates_brazil['Recovered']),go.Bar(name = 'Active Cases', x = topstates_brazil['State'], y = topstates_brazil['Active']),go.Bar(name = 'Death Cases', x = topstates_brazil['State'], y = topstates_brazil['Deaths']) ]) fig6.update_layout(title = 'Most Affected States in Brazil', barmode = 'stack', height = 600) fig6.show() # India fig7 = go.Figure(data = [go.Bar(name = 'Recovered Cases', x = topstates_india['State'], y = topstates_india['Recovered']),go.Bar(name = 'Active Cases', x = topstates_india['State'], y = topstates_india['Active']),go.Bar(name = 'Death Cases', x = topstates_india['State'], y = topstates_india['Deaths']) ]) fig7.update_layout(title = 'Most Affected States in India', barmode = 'stack', height = 600) fig7.show() # Russia fig8 = go.Figure(data = [go.Bar(name = 'Recovered Cases', x = topstates_russia['State'], y = topstates_russia['Recovered']),go.Bar(name = 'Active Cases', x = topstates_russia['State'], y = topstates_russia['Active']),go.Bar(name = 'Death Cases', x = topstates_russia['State'], y = topstates_russia['Deaths']) ]) fig8.update_layout(title = 'Most Affected States in Russia', barmode = 'stack', height = 600) fig8.show()# 2. Time series of top affected countries # We need a new data for this plot # https://covid19.who.int/ follow the link for this link for the next dataset(you can find the download option on the bottomright of the map chart) time_series = pd.read_csv('who_data.csv', encoding = 'ISO-8859-1') time_series['Date_reported'] = pd.to_datetime(time_series['Date_reported'])# a. Covid-19 cases worldwide # Firsty Data time_series_dates = time_series.groupby('Date_reported').sum()# Let's Plot # Cumulative cases fig11 = go.Figure() fig11.add_trace(go.Scatter(x = time_series_dates.index, y = time_series_dates[' Cumulative_cases'], fill = 'tonexty',line_color = 'blue')) fig11.update_layout(title = 'Cumulative Cases Worldwide') fig11.show() # Cumulative death cases fig12 = go.Figure() fig12.add_trace(go.Scatter(x = time_series_dates.index, y = time_series_dates[' Cumulative_deaths'], fill = 'tonexty',line_color = 'red')) fig12.update_layout(title = 'Cumulative Deaths Worldwide') fig12.show() # Daily new cases fig13 = go.Figure() fig13.add_trace(go.Scatter(x = time_series_dates.index, y = time_series_dates[' New_cases'], fill = 'tonexty',line_color = 'gold')) fig13.update_layout(title = 'Daily New Cases Worldwide') fig13.show() # Daily death cases fig14 = go.Figure() fig14.add_trace(go.Scatter(x = time_series_dates.index, y = time_series_dates[' New_deaths'], fill = 'tonexty',line_color = 'hotpink')) fig14.update_layout(title = 'Daily Death Cases Worldwide') fig14.show()# b. Most Affected Countries over the time # Data # USA time_series_us = time_series[' Country'] == ('United States of America') time_series_us = time_series[time_series_us] # Brazil time_series_brazil = time_series[' Country'] == ('Brazil') time_series_brazil = time_series[time_series_brazil] # India time_series_india = time_series[' Country'] == ('India') time_series_india = time_series[time_series_india] # Russia time_series_russia = time_series[' Country'] == ('Russia') time_series_russia = time_series[time_series_russia] # Peru time_series_peru = time_series[' Country'] == ('Peru') time_series_peru = time_series[time_series_peru]# Let's plot # Cumulative cases fig15 = go.Figure() fig15.add_trace(go.Line(x = time_series_us['Date_reported'], y = time_series_us[' Cumulative_cases'], name = 'USA')) fig15.add_trace(go.Line(x = time_series_brazil['Date_reported'], y = time_series_brazil[' Cumulative_cases'], name = 'Brazil')) fig15.add_trace(go.Line(x = time_series_india['Date_reported'], y = time_series_india[' Cumulative_cases'], name = 'India')) fig15.add_trace(go.Line(x = time_series_russia['Date_reported'], y = time_series_russia[' Cumulative_cases'], name = 'Russia')) fig15.add_trace(go.Line(x = time_series_peru['Date_reported'], y = time_series_peru[' Cumulative_cases'], name = 'Peru')) fig15.update_layout(title = 'Time Series of Most Affected countries"s Cumulative Cases') fig15.show() # Cumulative death cases fig16 = go.Figure() fig16.add_trace(go.Line(x = time_series_us['Date_reported'], y = time_series_us[' Cumulative_deaths'], name = 'USA')) fig16.add_trace(go.Line(x = time_series_brazil['Date_reported'], y = time_series_brazil[' Cumulative_deaths'], name = 'Brazil')) fig16.add_trace(go.Line(x = time_series_india['Date_reported'], y = time_series_india[' Cumulative_deaths'], name = 'India')) fig16.add_trace(go.Line(x = time_series_russia['Date_reported'], y = time_series_russia[' Cumulative_deaths'], name = 'Russia')) fig16.add_trace(go.Line(x = time_series_peru['Date_reported'], y = time_series_peru[' Cumulative_deaths'], name = 'Peru')) fig16.update_layout(title = 'Time Series of Most Affected countries"s Cumulative Death Cases') fig16.show() # Daily new cases fig17 = go.Figure() fig17.add_trace(go.Line(x = time_series_us['Date_reported'], y = time_series_us[' New_cases'], name = 'USA')) fig17.add_trace(go.Line(x = time_series_brazil['Date_reported'], y = time_series_brazil[' New_cases'], name = 'Brazil')) fig17.add_trace(go.Line(x = time_series_india['Date_reported'], y = time_series_india[' New_cases'], name = 'India')) fig17.add_trace(go.Line(x = time_series_russia['Date_reported'], y = time_series_russia[' New_cases'], name = 'Russia')) fig17.add_trace(go.Line(x = time_series_peru['Date_reported'], y = time_series_peru[' New_cases'], name = 'Peru')) fig17.update_layout(title = 'Time Series of Most Affected countries"s Daily New Cases') fig17.show() # Daily death cases fig18 = go.Figure() fig18.add_trace(go.Line(x = time_series_us['Date_reported'], y = time_series_us[' New_deaths'], name = 'USA')) fig18.add_trace(go.Line(x = time_series_brazil['Date_reported'], y = time_series_brazil[' New_deaths'], name = 'Brazil')) fig18.add_trace(go.Line(x = time_series_india['Date_reported'], y = time_series_india[' New_deaths'], name = 'India')) fig18.add_trace(go.Line(x = time_series_russia['Date_reported'], y = time_series_russia[' New_deaths'], name = 'Russia')) fig18.add_trace(go.Line(x = time_series_peru['Date_reported'], y = time_series_peru[' New_deaths'], name = 'Peru')) fig18.update_layout(title = 'Time Series of Most Affected countries"s Daily Death Cases') fig18.show()# 3. Case Classification and Distribution# For this we need a new dataset # https://www.kaggle.com/imdevskp/corona-virus-report follow this link for the next dataset# a. WHO Region-Wise Distribution # For this plot we are going to use country_wise_latest dataset which will come along with the downloaded kaggle dataset # Firstly Data who = pd.read_csv('country_wise_latest.csv') who_region = pd.DataFrame(who.groupby('WHO Region')['Confirmed'].sum()) labels = who_region.index values = who_region['Confirmed'] # Let's Plot! fig9 = go.Figure(data=[go.Pie(labels = labels, values = values, pull=[0, 0, 0, 0, 0.2, 0])]) fig9.update_layout(title = 'WHO Region-Wise Case Distribution', width = 700, height = 400, margin = dict(t = 0, l = 0, r = 0, b = 0)) fig9.show()# b. Most Affected countries case distribution # For this plot we are going to use the same country_wise_latest dataset# Firstly Data case_dist = who # US dist_us = case_dist['Country/Region'] == 'US' dist_us = case_dist[dist_us][['Country/Region','Deaths','Recovered','Active']].set_index('Country/Region') # Brazil dist_brazil = case_dist['Country/Region'] == 'Brazil' dist_brazil = case_dist[dist_brazil][['Country/Region','Deaths','Recovered','Active']].set_index('Country/Region') # India dist_india = case_dist['Country/Region'] == 'India' dist_india = case_dist[dist_india][['Country/Region','Deaths','Recovered','Active']].set_index('Country/Region') # Russia dist_russia = case_dist['Country/Region'] == 'Russia' dist_russia = case_dist[dist_russia][['Country/Region','Deaths','Recovered','Active']].set_index('Country/Region')# Let's Plot! # This plot is produced with matplotlib fig = plt.figure(figsize = (22,14)) colors_series = ['deepskyblue','gold','springgreen','coral'] explode = (0,0,0.1)plt.subplot(221) plt.pie(dist_us, labels = dist_us.columns, colors = colors_series, explode = explode,startangle = 90,autopct = '%.1f%%', shadow = True) plt.title('USA', fontsize = 16)plt.subplot(222) plt.pie(dist_brazil, labels = dist_brazil.columns, colors = colors_series, explode = explode,startangle = 90,autopct = '%.1f%%',shadow = True) plt.title('Brazil', fontsize = 16)plt.subplot(223) plt.pie(dist_india, labels = dist_india.columns, colors = colors_series, explode = explode, startangle = 90, autopct = '%.1f%%',shadow = True) plt.title('India', fontsize = 16)plt.subplot(224) plt.pie(dist_russia, labels = dist_russia.columns, colors = colors_series, explode = explode, startangle = 90,autopct = '%.1f%%', shadow = True) plt.title('Russia', fontsize = 16)plt.suptitle('Case Classification of Most Affected Countries', fontsize = 20)# c. Most affected continents' negative case vs positive case percentage composition # For this we need a new dataset # https://ourworldindata.org/coronavirus-source-data Follow this link for our next dataset# Firstly Data negative_positive = pd.read_csv('owid-covid-data.csv') negative_positive = negative_positive.groupby('continent')[['total_cases','total_tests']].sum() explode = (0,0.1) labels = ['Postive','Negative'] colors = ['beige','paleturquoise']#Let's Plot! fig = plt.figure(figsize = (20,20)) plt.subplot(321) plt.pie(negative_positive[negative_positive.index == 'Asia'],labels = labels, explode = explode, autopct = '%.1f%%', startangle = 90, colors = colors, shadow = True) plt.title('Asia', fontsize = 15)plt.subplot(322) plt.pie(negative_positive[negative_positive.index == 'North America'],labels = labels, explode = explode, autopct = '%.1f%%', startangle = 90, colors = colors, shadow = True) plt.title('North America', fontsize = 15)plt.subplot(323) plt.pie(negative_positive[negative_positive.index == 'South America'],labels = labels, explode = explode, autopct = '%.1f%%', startangle = 90, colors = colors, shadow = True) plt.title('South America', fontsize = 15)plt.subplot(324) plt.pie(negative_positive[negative_positive.index == 'Europe'],labels = labels, explode = explode, autopct = '%.1f%%', startangle = 90, colors = colors, shadow = True) plt.title('Europe', fontsize = 15)plt.suptitle('Continent-Wise Tested Positive & Negative Percentage Composition', fontsize = 20)翻譯自: https://medium.com/swlh/covid-19-analysis-with-python-b898181ea627
用python進(jìn)行營銷分析
總結(jié)
以上是生活随笔為你收集整理的用python进行营销分析_用python进行covid 19分析的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。
- 上一篇： 做梦梦到闪电是什么意思
- 下一篇： python实训英文_GitHub -