Since I started data analysis when I changed jobs in October 2020, I wanted to learn systematically, so I worked on the ** Python3 Engineer Certification Data Analysis Exam **. It was a relatively new exam, and it was a little different from the information on the web, so I will share my experience. I hope it will help those who are going to take the exam.
I will also write about myself, who is writing this article. I think it would be nice if you could think of this as a human impression. ** Nothing is more unreliable than "easy" as a clever person says. Lol**
--Two years of experience as an IT engineer --Most of the carriers are web-based backend development using Golang --Python history is about the last month after changing jobs
--Graduated from technical high school ――Since the curriculum is different from that of ordinary high school, there are times when you don't learn it in the first place. --F Run University dropout (Faculty of Non-Information Systems) ――Probably ** a person with very poor mathematical education **.
--Residing in Osaka
――As I aimed, I'm glad that I was able to study data analysis using Python firmly and systematically. ――It was good to deepen your understanding of "standard" libraries such as NumPy and pandas. ――It was the first time to use a library of machine learning and graph drawing, but it was good to learn from the basics. ――Compared to this test, the difficulty level of the mock test (described later) is considerably higher, and I was sick until I took this test. ――Personally, I found the foundations of mathematics difficult. ――During the test, I was told that calculation and memo paper were prohibited, and I had to solve it by mental arithmetic, which was difficult to solve. ――It was a system in which the application for the examination was contacted directly to the examination venue, and I felt the trouble of going to pay directly at the time of application.
** The main teaching material for this exam is Shoeisha's "New Data Analysis Textbook with Python" **. Basically, questions can only be asked from here. On the official website, questions were given in the following range and ratio.
|Chapter th>||Section th>||Number of problems th>||Problem rate th>|
|1||Role of data engineer td>||2||5.0%|
|2||Python and environment td>|
|1||Execution environment construction td>||1||2.5%|
|2||Python basics td>||3||7.5%|
|3||Foundations of Mathematics td>|
|1||Basic knowledge for reading mathematical formulas td>||1||2.5%|
|2||Linear algebra td>||2||5.0%|
|3||Basic analysis td>||1||2.5%|
|4||Probability and statistics td>||2||5.0%|
|4||Library analysis practice td>|
|5||Application: Data collection and processing td>||0||0%|
As shown in the table, the usage of ** 4 libraries closed 67.5% ** of the points, and this was clearly the important point.
The following is what I learned when taking this exam.
I took a quick look at the teaching materials and checked the operation with Jupyter Notebook in Chapter 4. Also, based on this teaching material, I created a question-and-answer question and solved it myself repeatedly. By the way, as mentioned above, ** this exam has a fairly biased range of questions **, so after learning the basics of mathematics to some extent, I decided to focus on ** studying the library **.
The above is a simple but experience report.
When studying, I also made my own problem, so I will publish it ↓ https://qiita.com/pon_maeda/items/39877a703b155d3dbdbd
If you have any questions, please feel free to comment. Thank you for reading until the end.