Kalan's Blog

Kalan 頭像照片,在淡水拍攝,淺藍背景

四零二曜日電子報上線啦!訂閱訂起來

Software Engineer / Taiwanese / Life in Fukuoka
This blog supports RSS feed (all content), you can click RSS icon or setup through third-party service. If there are special styles such as code syntax in the technical article, it is still recommended to browse to the original website for the best experience.

Current Theme light

我會把一些不成文的筆記或是最近的生活雜感放在短筆記,如果有興趣的話可以來看看唷!

Please notice that currenly most of posts are translated by AI automatically and might contain lots of confusion. I'll gradually translate the post ASAP

The Journey to Data Science from Scratch

Introduction

It's rare to have a relatively complete machine learning and data analysis team in the company, along with a comprehensive set of resources and pipelines. Taking advantage of this opportunity, I can learn about data science while also seeking help from colleagues if there are any areas I don't understand. The field of data science is vast, and currently, there is no clear direction. So, let's learn and explore simultaneously to see what sparks we can ignite. Embrace the slash!

Plan

  • Piece together the missing puzzle from university statistics course (2 weeks)

    • Distributions and various testing methods
    • R: Learn while reviewing
  • Review linear algebra (3 weeks)

    • I've completely forgotten everything
  • Review Andrew Ng's machine learning course (1 week)

    • I took this course using Octave, but the concepts should be similar
  • Complete Coursera's Deep Learning Specialization (6 weeks)

  • Finish reading the data science books I previously purchased (3 weeks)

  • Follow the progress of fast.ai and simultaneously delve into research papers

    • I previously completed chapters 1 to 4, but I've forgotten everything
  • Data engineering

    • Airflow
    • Kafka
    • Personally interested in learning these two

This should take approximately 100 days. Maybe I'll also take some data analysis-related courses to further enhance my skills. Currently, it seems that machine learning and deep learning are dominant. In any case, I'll record what I've come up with here to avoid forgetting.

Goals

The main objective is to explore the exciting aspects of integrating front-end development with data science, such as working with ml.js or tensorflow.js, which seem fascinating.

Additionally, many of my ideas require the assistance of data science, so I'll take advantage of the current availability of time to supplement my knowledge in this area. My previous notes were scattered all over the place and are almost impossible to find now. Moreover, I've forgotten a large portion of what I learned. This time, I will make sure to document everything properly on my blog.

Prev

Project Winter play experience

Next

Implementing Air Quality Monitoring Applications with Arduino and ESP32 (1) - Introduction to Sensors

If you found this article helpful, please consider buy me a drink ☕️ It'll make my ordinary day shine✨

Buy me a coffee