[PYTHON] How to study deep learning G test

How to study deep learning G test

I took the ** Deep Learning G (Generalist) Test # 3 ** that was held on November 9, 2019 and passed it successfully, but I would like to summarize the study method.

What is G test?

・ Purpose: To test whether you have the knowledge to utilize deep learning in your business ・ Eligibility restrictions: None ・ Exam outline: 120 minutes, knowledge questions (multiple choice), online (home exam) ・ Question: Questions from the syllabus

Purpose of the exam

My main job is a production engineering engineer, and I also work in data science, etc., and also manufacture data analysis and image recognition systems using machine learning and deep learning. I am studying with books, but I took the exam because I wanted to systematically relearn. By getting a qualification, I also intended to show that I have the will and ability to focus on my future work.

Study period

Start studying for the exam 3 weeks before the exam.

However, I have read many AI-related books from the beginning, and also have experience in implementing machine learning and deep learning. I have been taking Udemy related courses for several years. Therefore, I thought I knew it, so I was a little off guard and decided to study for the exam at once just before. When I studied, there were many recommended books, and I didn't have enough time to study well. If you just want to pass the exam, three weeks is enough.

content of study

I proceeded with the study in the following books. ① "Deep Learning Textbook Deep Learning G Test (Generalist) Official Text" ② "Thorough capture Deep learning G test generalist problem collection" ③ Recommended books "AI White Paper 2019", "Textbook for Deep Learning Utilization", etc. ④ kindle "G test to understand in practice Deep learning textbook: Secret method to pass in the shortest time taught by G test passers (Kamikusa Publishing)" "Detailed explanation! G test to understand in practice Web mock test manual"

Especially, ①②④ was repeated about 3 times. ③ is a level that I haven't had enough time to read. Especially, the web mock test attached to ④ was good.

Take the exam

・ There are many legal issues unexpectedly ・ There are quite a few problems that are not in the text These survived Google search

By repeating ①②④ of the above book three times, I was able to solve the basic problem at the speed of instant killing, so I solved all the questions with about 15 to 20 minutes left, and Google solves the unconfident problems one by one. I have time to search and check with. As for the sense of speed of the test, it was good that I was able to experience the test in a very similar format in advance by doing the web mock test in (4) above (the web mock test is easy, it takes more than 30 minutes, and 90% or more can be taken. ). However, I have the impression that the actual exam is more difficult (many unknown questions, especially legal questions).

For the time being, if you have any questions that you do not understand during the exam, select an answer and proceed quickly. It is effective to search for the remaining time and make sure to get it. Due to the large number of questions, it was important not to spend time on the basic questions, which helped with repeated training. I think there was a problem as it was in the problem collection.

After all, it was a legal issue that was difficult to attack with common sense, so I think it is worth reading through the latest books such as the "AI White Paper". To be honest, it's difficult to study here, so even if you look it up on Google, you can pass it if you have confidence in the basics. It seems that the pass line is not open to the public, but if you study repeatedly based on textbooks and problem books, you can aim for basic and technical problems almost perfectly, so I think that you can pass enough.

Since a wide range of knowledge is required, I think I was able to systematically understand what I had read and learned in various books and organize my mind. To be honest, I'm glad I received it. It's a field that is evolving steadily, so I would like to collect information and update my knowledge again.

Next is E qualification

Now that I have passed the G test, I will challenge the ** E (engineer) qualification **, which requires mounting ability. There will be an exam in February 2020 in the near future, so there are only about 3 months left. Moreover, it is necessary to complete the course of the programming school specified for the examination qualification. As a result of the selection, I am taking an AI job curry course at COSPA (which was even cheaper just before). I am receiving it and it progresses at a considerable speed, so honestly it is a tough impression for beginners such as python. For the time being, I can use python and have experience in data science, so I can keep up with it, but it will be a fairly tight schedule to pass the completion exam of 3 courses in the last 3 months and take the E qualification exam. Even so, I have knowledge in the G test, so I would like to take this opportunity to get the E qualification at once. I will do my best in a short period of time!

Recommended Posts

How to study deep learning G test
How to study for the Deep Learning Association G test (for beginners) [2020 version]
How to study Bukkit Plugin
Introduction to Deep Learning ~ Learning Rules ~
Deep Reinforcement Learning 1 Introduction to Reinforcement Learning
Introduction to Deep Learning ~ Backpropagation ~
[Django] How to test Form [TDD]
Introduction to Deep Learning ~ Function Approximation ~
Deep learning to start without GPU
Introduction to Deep Learning ~ Coding Preparation ~
Introduction to Deep Learning ~ Dropout Edition ~
Introduction to Deep Learning ~ Forward Propagation ~
Thoroughly study Deep Learning [DW Day 0]
Introduction to Deep Learning ~ CNN Experiment ~
How to collect machine learning data
Deep Learning
Introduction to Machine Learning: How Models Work
scikit-learn How to use summary (machine learning)
Reinforcement learning to learn from zero to deep
I installed Python 3.5.1 to study machine learning
Introduction to Deep Learning ~ Convolution and Pooling ~
How to use Google Test in C
How to do portmanteau test with python
Image alignment: from SIFT to deep learning
How to enjoy Coursera / Machine Learning (Week 10)
How to test on a Django-authenticated page
[Python] How to test command line parser click
Deep Learning from scratch ① Chapter 6 "Techniques related to learning"
Study from Python Hour7: How to use classes
Start Deep learning
Python Deep Learning
Introduction to Deep Learning ~ Localization and Loss Function ~
Deep learning × Python
[For beginners] How to study programming Private memo
How to install the deep learning framework Tensorflow 1.0 in the Anaconda environment of Windows
How to study Python 3 engineer certification data analysis test by Python beginner (passed in September 2020)
How to perform learning in SageMaker without session timeout
[Deep Learning Association E Qualification] What to do to receive
Learning Deep Forest, a new learning device comparable to DNN
Artificial intelligence, machine learning, deep learning to implement and understand
[Deep Learning from scratch] I tried to explain Dropout
[TF] How to save and load Tensorflow learning parameters
How to use xml.etree.ElementTree
How to use Python-shell
First Deep Learning ~ Struggle ~
How to use tf.data
How to use virtualenv
Scraping 2 How to scrape
How to use image-match
How to use shogun
How to install Python
How to use Pandas 2
Python: Deep Learning Practices
Deep learning / activation functions
How to read PyPI
How to install pip
How to use Virtualenv
[Deep learning] Investigating how to use each function of the convolutional neural network [DW day 3]
How to use numpy.vectorize
How to update easy_install
How to install archlinux