STEM

Data science project: machine learning

0 Flares Twitter 0 Facebook 0 Pin It Share 0 Filament.io 0 Flares ×

This week, I was sick and had to stay home. This was of course bad but this gave me the opportunity both to spend more time with the children (also sick at home) and to work on my data science course. And I finally completed the course 8: Machine learning with Python

Course 8: Machine learning with Python

I completed the course on December 10th 2019. This was an interesting course. You learn for instance how recommendation systems work on popular websites such as Amazon or Netflix.

The part on regression was not totally new. I had already seen these notions in previous courses. However, the parts on classification, clustering and recommendations included a lot of new information and techniques and it was a lot of work.

The final assignment focused mainly on classification so this is the part of the course which I master best now. However, I found the questions in the assignment not always quite clear, and you have to check the course forum to understand really what is expected.

IBM data science progress

Course 9: Applied data science capstone

This is the very last course and it will enable me to work on a “real life” data science project. I am curious about how difficult it will be. According to the course description, it should take 5 weeks of studies or a total of 45 hours. I found that the expected time of completion for other courses was not always accurate though, so I will see how long it will really take me. When I complete the course, I will receive a IBM data science professional certificate.

Related posts

You can see below my previous posts on the project:

4 Comments

  • matthewrbaker

    Congratulations on being almost done with the course! Data science is something I plan to learn next year. Do you recommend this course? Any other resources you would suggest?

    • momslovelearning

      That’s great! I think you will enjoy data science, this is a very interesting field.
      I found this course very good. The great things about the courses are:
      – it is very accessible for beginners. If you work consistently, you will be able to finish it.
      – you can choose either to follow the courses for free or to pay for a certificate.
      – it covers most of the fields a data scientist is confronted with
      The things that could be improved:
      – the assignment grading by peers is not perfect. It may be that the person grading your assignment does not have a great understanding of the field. Moreover, the assignments are not always clear, which can bring confusion.
      – for some of the elements you learn, there are too few exercises. The final assignments often focus on one or two chapters of a course, which means that you may want to work further on the other points on your own.
      As for other resources, I would also recommend to have a look on the Codecademy website. They have a paying data science track that I have not tried and they have free introduction courses to the main programming languages that are used in data science: Python and R as well as SQL. I had taken the free courses on Python and SQL before I started the course and both helped me to go faster through the IBM course.
      By Coursera, there is also a data science with Python specialization from the University of Michigan. As far as I could see, it is very good but it is difficult if you do not have already experience with Python (especially pandas). So if you are a beginner, the IBM course is better and if you have already some basis, the University of Michigan course may be better.
      There are also good data science courses on Coursera that are focusing on R or on SQL.
      There are also some resources on LinkedIn. Checking the data science jobs that are proposed there will help you to see which skills are expected and you can also follow data scientists that publish articles or video on their job.

      • matthewrbaker

        Wow, great info! Thank you! I did learn Python and SQL in college but haven’t had to use them in a few years. I will be sure to look into the IBM course, since it sounds like a decent starting point. And I’ll check out Coursera and LinkedIn as well.
        I just remembered that there is also a site called Lynda that may have data science courses. I have free access through my city library; you might be able to get access in the same way. I’ll check it out too, but just thought I’d pass the info along!
        And you are right about checking the required skills for data science jobs. I’ve looked a little bit and seen that Python, SQL, and R are all pertinent. Some also would like experience with Tableau or Excel. But I think in many cases that if one has two or three skills (especially the programming ones), an employer would be willing to let one learn the other skills on the job.
        Thanks again!

        • momslovelearning

          Thanks for the info! I will check the courses on Lynda. The IBM course is on Coursera, and maybe you can get free access there too. I could get a three months Coursera license through the company I am working for, for instance.
          As you have already learned Python and SQL in college, you should find most of the data science beginner courses easy to go through. If you are using Coursera, you should probably wait before you buy a certificate and try a few courses on the free modus to see which one is better adapted to you.

Leave a Reply

0 Flares Twitter 0 Facebook 0 Pin It Share 0 Filament.io 0 Flares ×
%d bloggers like this: