@Takp Blog

I just completed CS 7641: Machine Learning course. It was my 4th course in the Georgia Tech OMSCS (Online Master of Science in Computer Science) program, and fortunately, I could earn the letter grade A! It was an exciting journey, but it required a lot of effort to complete the assignments and to study for the exams. I’m gonna explain what I’ve done and what I thought about this course. 1. [Read More]

When you need to calculate the modular inverse for the large co-prime integers, how do you calculate? In my case, I needed to calculate the modular inverse to get the RSA private key from the given public exponent and RSA modulus. At first, I tried to find it by brute-force search, but it turns out it takes very long time to compute the modular inverse for the large numbers. Finally, It is possible to calculate modular inverse efficiently using extended GCD function. [Read More]

1. Needs for better data serialization Now I’m building web applications that are running on Docker containers. Each containers need to interact with the others. Then, which data format is the best to communicate among them? Nowadays most of web APIs are using JSON with REST api under HTTP protocol. JSON and XML are readable for humans and it is big advantage of them. But when it comes to the data size, JSON and XML are not good. [Read More]

1. Accepted to Georgia Tech OMSCS! September 2018, I received official acceptance letter from Georgia Tech OMSCS - Online Master of Science in Computer Science! 🎉 This OMSCS program is an online course, but the screening process is exactly same to on-campus program. So it is required to prepare SOP, recommendations and TOEFL score. Here I’m going to explain what’s needed in this article. 2. Admission Criteria It requires: Bachelor’s degree GPA 3.0 or higher If applicants do not meet the criteria, it will be judged by case-by-case. [Read More]

Algorithms course by Stanford Univeristy I could complete the algorithms course which I started from April 2018! 🎉 It took me about 4 months to finish. This course is one of the Massive Open Online Courses (so-called “MOOCs”), and is hosted by Coursera. It’s open with the title “Algorithms, a 4-course specialization by Stanford University” and the classes are all made by Stanford Univeristy. This course is composed of 4 courses and you can complete all courses within 4 months. [Read More]

The paper about image inpainting was presented to SIGGRAPH, and I give it a try since the actual model was published. Paper: “Globally and Locally Consistent Image Completion” Here this is the original paper presented to SIGGRAPH. Globally and Locally Consistent Image Completion Take a look at the Github This github repository was created by the author of the paper. It seems to be published from around February 2018. Github - https://github.com/satoshiiizuka/siggraph2017_inpainting We can download the trained model from this github. [Read More]

1. Why I implemented it in Ruby I created simple version of blockchain in Ruby. This is because I read this article “Learn Blockchains by Building One” and I’m impressed with this. It creates simple blockchain in Python less than 200 lines of code. Then I thought why not in Ruby? For the engineers, it is easier to figure out something by creating simple one by coding, isn’t it? 2. Implement blockchain in Ruby Just implement it in Ruby one by one. [Read More]

Udacity’s Machine Learning Nano-degree (MLND) I’ve graduated the Udacity’s Machine Learning Nano-degree course! Udacity is known as one of the best MOOC. It takes about half year from June to December 2017. This is the link to the course I graduated. Udacity - Machine Learning Engineer Nanodegree Why I studied at Udacity after I completed the Coursera’s machine learning course I started to study Udacity after I completed the Coursera’s Machine Learening course (article). [Read More]

In April, I contributed to Mastodon, which became so popular as twitter-like web service. This is how I contributed. How I start It was so getting popular at that time, and I was watching the source code of this project because this was made by Rails & React. (Mastodon Github) I was also watching the issues, then I found the bug report that it failed the internationalization (I18n) on error pages like 404, 500 pages. [Read More]

I could completed the Machine Learning course by Stanford University! Yay! I started in April, and I could finish it at the end of June. So It took about 3 months. (Please refer to the entry, which was written in the beginning => Started to study Coursera Machine Learning course by Stanford Univ. ) Cons Unfamiliar programming language “Octave”. But Octave is quite intuitive for calculation. A bit hard to submit the programming assignment every week, but this became good training. [Read More]