How to study Python 3 engineer certification data analysis test by Python beginner (passed in September 2020)

This time, I passed the "Python3 Engineer Certification Data Analysis Test", so I will describe how to learn it.

■ Author information (as of August 2020)

・ Infrastructure engineer. ・ AWS qualifications have been acquired by professionals. ・ "G test" has passed (taken on July 4, 2020).

■ Motivation for taking an examination

It all started when I tried to get an AWS expertise qualification earlier this year.

I thought that AI-related things that might affect business in the future would be good, so I aimed for "AWS Certified Machine Learning". However, even though I had knowledge of AWS, I had no knowledge of AI or machine learning. Therefore, I first obtained the G test, but I learned the importance of "Python" in machine learning and obtained the "Python3 engineer certification basic test". I took the "Python3 Engineer Certification Data Analysis Exam" this time.

■ Exam contents

The contents of the test are as follows.

Examination fee: General: 10,000 yen (excluding tax)

Teaching material name Classification information
Python 3 Engineer Certified Data Analysis Exam(official) Web(free) まずはofficialサイトで試験内容を確認しました。この試験の特徴はテスト範囲が「Pythonによるあたらしいデータ分析の教科書」(翔泳社)から出題されることが明確になっている点です。どの章から何問出題されるか、まで明確になっています。例えば5章からは1問も出題されません。

■ Study methods and teaching materials used

The study methods and teaching materials used are as follows.

Teaching material name Classification How to study
[Official text]
A new data analysis textbook with Python(official)
kindle(Paid) すべての出題がここからされる前提となっているofficialテキストです。

Since it is described as a "textbook", it may be misunderstood if a beginner reads it, but in reality it is not described for beginners at all. At the beginning of the book, there is a statement that "the textbook presents the necessary information necessary for data analysis and explains it concisely", but the explanation is too concise and even beginners can not understand it due to lack of explanation. Because.

In order to understand the content, it is necessary to search the keywords in the book online or supplement them with other teaching materials. I feel that it is more of a "book for clarifying the test range" than a "book for studying".

[Learn in one day] Python data analysis for engineers Web(Paid) This is a Udemy course. Usually 10,It costs 800 yen, but 90 during the time sale period, which is held about once every three months.%Close discount 1,It can be purchased for about 400 yen. I myself bought it at a time sale.

The video touched on installing Anaconda, using Spyder, Jupyter notobook, and basic Python operations, and was very easy to understand. The chapters before Numpy are also recommended for those who take the "Python3 Engineer Certification Basic Exam". However, in the later chapters, "Matplotlib" and "scipy" will suddenly become difficult, so I think it's okay to take a look at those areas and exercises at first.

Considering the test range of the data analysis test, "Foundation of Mathematics" and "scikit"-There is a lack of "learn", which needs to be supplemented in some other way.

After all the ease of understanding the video is the best.

Learn at the prep school Web(free) 「数学の基礎」の部分はWebで調べるより動画で勉強した方が理解しやすいことを「G検定」の学習の中で学びました。「Learn at the prep school」はその中でも特におすすめの動画になります。もちろん他の講師の方の動画もYoutubeに多数アップされているので、自分が理解しやすい動画を探すことが大事です。

■ PRIME STUDY practice test Web(free) This collection of questions is the key to passing this exam. The official text is so wide-ranging that I have 100 about the content and keywords that appear in these three practice exams.%Emphasis was placed on understanding. Although it is stated in the official textbook as my study method, I will not be able to solve the problem that the keyword does not appear in the mock exam, but I divided it and bet on the mock question collection.

■ PRIME STUDY practice test implementation details (reference)

I thoroughly understood the contents of the mock question collection in the first time, and then repeated it. The last week of study was to solve this mock exam once a day. The following is the execution result.

Practice test name Score Pass / Fail Implementation date result
1st practice test 047.5 points failure August 30, 2020 After one lap of the textbook. The situation that I do not understand at all. Learn to fully understand all problems.
1st practice test 100.0 points Pass September 06, 2020 Mock exam content After reviewing, retake the exam.
2nd practice test 080.0 points Pass September 06, 2020 なんとか初回でPass点超える。全部の問題を完全に理解できるように学習する。
2nd practice test 100.0 points Pass September 06, 2020 Mock exam content After reviewing, retake the exam.
3rd practice test 085.0 points Pass September 06, 2020 初回でPass点超える。全部の問題を完全に理解できるように学習する。
3rd practice test 100.0 points Pass September 06, 2020 Mock exam content After reviewing, retake the exam.
1st practice test 100.0 points Pass September 07, 2020 Implemented once a day as a target.
2nd practice test 100.0 points Pass September 08, 2020 Implemented once a day as a target.
3rd practice test 100.0 points Pass September 09, 2020 Implemented once a day as a target.
1st practice test 100.0 points Pass September 10, 2020 Implemented once a day as a target.
2nd practice test 100.0 points Pass September 11, 2020 Implemented once a day as a target.
3rd practice test 097.5 points Pass September 12, 2020 One question is wrong due to a general mistake.

Terms of Service has the following description, so I cannot mention the contents. Also, regarding the email address, it is safer to use an address that does not cause any problems even if the email is delivered.

  1. Information, photographs and other copyrighted works (hereinafter referred to as "literary works") posted on this website by the Company shall belong to the Company or the author or copyright holder who created the works. ..
  1. The user shall not infringe or may infringe the copyright of our work, such as copying, public transmission, transfer, adaptation and translation.
  1. The user agrees that advertisements, etc. will be posted on this service and that the e-mail newsletter containing the advertisements, etc. will be delivered. The Company shall not be liable for any damages caused by transactions conducted by advertisements, etc. posted on this service or e-mail newsletter, and damages caused by the advertisement itself.
  1. The user agrees that we will send you an email regarding recruitment. We will not disclose any standards for email delivery related to recruitment.

■ Exam (September 12 exam: Pass)

point Contents
Examination hall Oddessey ID & PasswordAnd don't forget the "designated ID".[Oddessey ID & Password]If you can't remember, the staff will allow you to bring in a sticky note. Other test takers besides me also received sticky notes.

I was allowed to go to the bathroom on the way during other tests, but this test was based on the premise that I would end when I went out, even if I was ill. Many times before the test started, I was prompted to go to the bathroom in advance, which put pressure on me. I was nervous when I took the exam if my stomach hurts on the way.

また、今回はPayPalで事前に受験料を支払いしたのですが、私のExamination hallでは当日に会社宛ての領収書を発行してもらえました。これがないと会社の事務処理が面倒くさいので助かりました。

Impressions The content of the problem cannot be described by the rules. It took about 30 minutes to solve 40 questions. However, it was dangerous because I was worried about answering about 12 out of 40 questions. 28 questions because it passed with 700 points or more(28 x 25 points = 700 points)Was a last-minute line that needed to be answered correctly and could be mistaken. However, all I didn't understand was memorization problems, and I couldn't help taking time, so I gave up and ended the test.

result テストが終了した後はアンケート画面に遷移します。アンケートの回答が完了すると合否が表示されます。私は「合格」で、正解率は725点でした。あと2問間違っていたら、不合格になるところでした。久々に薄氷のresultです。


■ Ideal study method for Python beginners

It is a point assuming the same Python beginner as me.

(1) Qualifications that should be obtained in advance

There are two qualifications that you should obtain in advance. There are the following two. If you have reached the level where you can acquire these two, I think that you will understand the qualification study of data analysis quickly.

Qualification Reason
G test 機械学習のことや手法などが学べます。公式テキストには、さらっと回帰・分類、クラスタリングといった用語が出てくるのですが、Gtestを取得していたおかげで、すんなり受け理れることができました。

Gtest合格者による受験に向けた心構えについて(2020年7月合格)I mentioned the article in ", but it is a qualification that can be obtained in a study period of about one month.

Python 3 Engineer Certification Basic Exam IfyouarenewtoPython,thisisdefinitelythewaytogo.IthinkthedataanalysistestisessentialbecauseitrequiresthatyouknowthebasicsofPython.

HowtostudyPython3engineercertificationbasicexambyPythonbeginner(passedinAugust2020)I mentioned the article in ", but it is a qualification that can be obtained in a study period of about one month.

(2) Study method using official textbooks

As mentioned earlier, "New data analysis textbook using Python (official) The official textbook is a difficult material for beginners to study first. Therefore, I think it is efficient to study in advance so that the contents of this teaching material will be in your mind.
Udemy "[Learn in one day] Python data analysis for engineers" that I used for learning was very effective, so please I recommend it. The video will teach you how to install Anaconda, how to use Spyder and Jupyter notobook, how to operate Python, and how to use libraries such as Numpy in an easy-to-understand manner. When I read the official textbook after learning this, I was able to understand it to some extent. I think it will lead to more efficient learning and shorter time. Regarding this pre-learning material, I think that the material that suits each individual is different, so if it does not suit you, please try to find the material that suits you.

★ Points for purchasing teaching materials at "Udemy"

① When purchasing teaching materials at "Udemy), please purchase them at the time sale held about once every three months. This teaching material usually costs 10,800 yen, but the price drops to around 1,400 yen only during the time sale period.
(2) Some videos are released as samples in the "Course content" part of the teaching materials. Please read this content and purchase it if you think it suits you. If it doesn't suit you, I think you should look for other teaching materials.

(3) Preparation method for the production test

As a preparation method for the production test, I think it is effective to carry out until you thoroughly understand "■ PRIME STUDY practice test". I will. It is important to study until you understand not only the correct answer, but also why the other options are wrong.

In addition, as a supplement to the mock questions when studying, all three mock questions are classified as follows. For example: 1. Problems related to the role of data engineer are always 1 and 2. If you know this premise, you can compare the contents of each mock question and find out where in the official text you can find the description.

Chapter number title The number of questions Problem rate Mock question question number
1 Role of data engineer 2 5.00% 1,2
2.1 Execution environment construction 1 2.50% 3
2.2 Python basics 3 7.50% 4,5,6
2.3 Jupyter Notebook 1 2.50% 7
3.1 Basic knowledge for reading mathematical formulas 1 2.50% 8
3.2 linear algebra 2 5.00% 9,10
3.3 Basic analysis 1 2.50% 11
3.4 Probability and statistics 2 5.00% 12,13
4.1 NumPy 6 15.00% 14,15,16,17,18,19
4.2 pandas 7 17.50% 20,21,22,23,24,25,26
4.3 Matplotlib 6 15.00% 27,28,29,30,31,32
4.4 scikit-learn 8 20.00% 33,34,35,36,37,38,39,40
5 application:Data collection and processing 0 0.00%

However, as stated in my test results, there is a possibility that you will pass the exam with just the right score. From my reflection, I think it's better to organize functions and methods into EXCEL or something and understand them to some extent. The more content you remember, the better your score will be and the more you will be able to pass.

■ Supplementary information on "New Data Analysis Textbook with Python (Official)"

I will describe what I was interested in while implementing the official text.

(1) Precautions when Anaconda is installed in advance

There is a procedure to install Python in "2.1 Execution environment construction". However, if you already have Anaonda installed, you do not need to install it. "In the Anaconda environment, pip is also available, so it is possible to use pip to add these packages. However, in rare cases, the environment built with the conda command may be corrupted." There is also a description that the Anaconda environment will be destroyed at worst, so let's stop. Even if you leave Anaconda as it is, there is no problem in the range up to Chapter 4 of the test range.

(2) About syntax execution described in the official text

When I buy the kindle version, paste the syntax and try to run it in python, I immediately get the following message: image.png

However, posting example sentences by hand takes an enormous amount of time. What should not be overlooked here is the sample data that describes the download site in the text. If you upload the folder "notobooks" downloaded here directly under the designated folder of "jupyter", you can open the page with the syntax of Chapter 4. Therefore, you can execute it immediately and check the operation. image.png

(3) How to install the "pydotplus" module

You need to install a module called "pydotplus" at "● Decision Tree" in "4.4.2 Classification". There was no description of the installation procedure by Anaconda, so I will describe it. Before doing this, let's stop "Jupyter" once. When you start "Jupyter" after the installation environment, the "pydotplus" module is available. ① Start Wincows "Start" button> Anaconda3 (64-bit)> Anaconda Navigator (anaconda3). ② Select "Environments". ③ Select "All". ④ Select "Update index" to update the index.

(4) About scikit-learn warning

An error will occur even if you write it according to the syntax with "scikit-learn". This is an error because some classes are obsolete. I was also in trouble, but there is a countermeasure on the page "I tried to deal with the scikit-learn warning I encountered during copying". ,It was helpful.

■ Impressions

It was great to be able to study the specific behavior of libraries such as Numpy, Pandas, Matplotlib, and scikit-learn that use Python. Even if I studied the library name in the G test, I could not understand the specific part such as what kind of processing is actually done. That question was resolved through this qualification study.

■ For the next time

We have been acquiring qualifications from "G test" ⇒ "Python basics" ⇒ "Python data analysis", but next is "E qualification" I would like to aim to acquire "AWS Certified Machine Learning".

I hope this article will serve as a reference for those who are taking the "Python3 Engineer Certification Data Analysis Exam".

that's all

Recommended Posts

How to study Python 3 engineer certification data analysis test by Python beginner (passed in September 2020)
How to study Python 3 engineer certification basic exam by Python beginner (passed in August 2020)
Have passed the Python Engineer Certification Data Analysis Exam
[For beginners] How to study Python3 data analysis exam
[Python3 engineer certification data analysis test] Examination / passing experience
How to pass and study the Python 3 Engineer Certification Basic Exam
Python3 Engineer Certification Data Analysis Exam Self-made Questions
Python 3 Engineer Certification Data Analysis Exam Pre-Exam Learning
How amateurs passed the Python 3 Engineer Certification Basic Exam
How an "amateur banker" passed the Python 3 Engineer Certification Basic Exam in a week
How to develop in Python
How to test that Exception is raised in python unittest
How to generate exponential pulse time series data in python
I passed the Python data analysis test, so I summarized the points
Data science companion in python, how to specify elements in pandas
(Maybe) This is all you need to pass the Python 3 Engineer Certification Data Analysis Exam
[Python] How to FFT mp3 data
[Python] How to do PCA in Python
A simple data analysis of Bitcoin provided by CoinMetrics in Python
How to collect images in Python
How to use SQLite in Python
I passed the python engineer certification exam, so I released the study method
How to use Mysql in python
How to wrap C in Python
How to use ChemSpider in Python
How to use PubChem in Python
How to create a large amount of test data in MySQL? ??
How to visualize where misclassification is occurring in data analysis classification
[Beginner memo] How to specify the library reading path in Python
How to handle Japanese in Python
"Introduction to data analysis by Bayesian statistical modeling starting with R and Stan" implemented in Python
How to send a visualization image of data created in Python to Typetalk
How do I represent the data passed in Curl --data-urlencode in Python Requests?
Data analysis: Easily apply descriptive and inference statistics to CSV data in Python
How to plot galaxy visible light data using OpenNGC database in python
How to study until a beginner in statistics gets started with Bayesian statistics
How to sort by specifying a column in the Python Numpy array.
How to deal with old Python versions in Cloud9 made by others
[Introduction to Python] How to use class in Python?
How to access environment variables in Python
How to dynamically define variables in Python
How to do R chartr () in Python
[Itertools.permutations] How to put permutations in Python
How to use Google Test in C
How to work with BigQuery in Python
How to study deep learning G test
How to get a stacktrace in python
How to display multiplication table in python
How to extract polygon area in Python
How to do portmanteau test with python
How to check opencv version in python
How to switch python versions in cloud9
How to adjust image contrast in Python
How to use __slots__ in Python class
How to dynamically zero pad in Python
How to use "deque" for Python data
How to use regular expressions in Python
How to display Hello world in python
Python 3 Engineer Certified Data Analysis Exam Preparation
How to use is and == in Python
How to write Ruby to_s in Python