good

RAS

-

Title: Frustrating Experience Due to Speaker and Technical Issues I enrolled in the course "Supervised Machine Learning: Regression and Classification" with high hopes, but I was left thoroughly disappointed. The most frustrating aspect of this course was the speaker's voice. At times, it was so annoying that it made it nearly impossible to concentrate on the content being presented. The speaker's tone and delivery lacked enthusiasm and clarity, making it difficult to stay engaged. To make matters worse, there were frequent microphone issues throughout the course. It felt like the words were either muffled or slurred, which not only made it hard to understand but also incredibly irritating to listen to. These technical problems seriously hindered my ability to learn and absorb the material effectively. While the course content itself was decent, the combination of the speaker's annoying voice and microphone problems made it a struggle to complete. I expected a much more professional and engaging learning experience, and sadly, this course fell far short of my expectations. Title: Valuable Content, Could Use Some Improvements I recently completed the course "Supervised Machine Learning: Regression and Classification," and I have mixed feelings about my overall experience. On the positive side, the course content was comprehensive and provided a solid foundation in supervised machine learning. The topics were well-structured, and I appreciated the depth of coverage on regression and classification techniques. The examples and exercises were helpful in reinforcing the concepts, and I did gain valuable insights into these subjects. However, there were some notable drawbacks. First and foremost, the speaker's voice occasionally made it challenging to stay engaged. There were moments when the tone and delivery were less than engaging, which detracted from the learning experience. Additionally, there were technical issues with the microphone that affected the audio quality. This made it difficult to understand certain parts of the course and was quite bothersome. In summary, the course content itself is valuable for anyone looking to learn about supervised machine learning, particularly regression and classification techniques. However, improvements in the speaker's delivery and addressing the technical issues with the microphone would greatly enhance the overall learning experience.

I'm a 45 year old software developer that never chose to progress in math beyond college algebra. I never needed it. when i took college algebra in highschool, most of the calculus related notation was not covered; this may have changed since the early 90s, but it isn't realistic to know where someone is sitting in regards to calculus. This course presented itself as if you would not need to know calc to keep up with the logic involved, but you really do need to have a good grasp on the notation involved. The frustrating thing for me is that I had to spend more mental energy just trying to digest the logic, and it really took away from the much more important content of knowing how and when to use the algorithms covered here. The math is already encapsulated in machine learning libraries, so the heavy work of trying to keep up was really mostly wasted. I think the course needs to be more up front about the math related background needed to step into it, or it should be taught with much more emphasis on applying the concepts involved in how and when to use the algorithms to create useful models. With more honesty about the background requirements to keep up, or with a shift in focus towards application, i would be giving this course a better review. but 3/5 or 3.5/5 is the best i can offer.

I enjoyed the way fundamentals were thought and it was comprehensive yet simple. I only have few complaints: 1. the practice lab and quizzes could be a bit harder and more twisted. and the number of programming assignments could be more 2. it will be great if the course focuses a little bit mode on building models using libraries such as scikit-learn rather than just explaining the theories and mathematics behind it. 3. it will be helpful to ass subjects such as cross validation, XGBoost, pipelines etc. that are essential to build an accurate model to make learners more familiar with the real word problems of training a model rather than just explaining the basics thanks for such a good content again.

I don't know if I'm perhaps not the target group of this course, so perhaps my review isn't relevant.

The course itself felt very slow and shallow, and did not give me a lot of theoretical background that I was interested in understanding.

Additionally, I find the optional labs, quite cumbersome to load and play with, additionally the practice quizzes were sometimes phrased a bit confusingly.

However, I do have to say that I understood the main ideas. It's just very not challenging.

The course provides a lot of very good high level information in Machine Learning (Regression and Classification) , but it feels like I was strolling through the park. The course lacks rigor, but makes up for it with very good explanations and intuition. Still leaves you wanting more. I think you are going to take up a students time, provide them with the same level of information as a graduate student. After all, they want to use this in the real world, do they not?

Not as good compared to the previous machine learning course (the one done with Octave), you aren't pushed to write the vectorised implementations, even the example code and starter code in assignments is done with loops. No introduction into the "@" syntax of numpy. I prefered the math syntax of the prev course with hypothesis function and Theta for weights, including adding the bias term itself as theta_0 within the weights vectors.

The educational material was good, but the assessments were very simple. The quizzes and assignments could have been more thorough to properly assess whether I had grasped the material, rather than "fill in the blanks" and multiple choice (though often only 2 choices) style problems. A certificate from this course is not a good indicator that the student has learned the material.

The course dives too deep into the math behind the type of regressions to make predictions. But it lacks practice in using python libraries to actually put them into use. Instead of having to calculate each operation manually it should present the learner with more opportunities to implement python machine learning libraries to get some hands on experience.

Error that was shown after an assignment is done was not helpful to debug on where the error is and took couple of hit and trial to work. The cell blocks said all tests successful but was unable to submit the assignment. User friendly text explanations on where the error is, Highlighting in red etc could have been helpful.

The videos were great and it was an excellent introduction to machine learning. However, I feel like the quizzes should test our knowledge a little more. The questions are too simple and makes it so that I am not sure if I actually have a good understanding of the material being taught.

Practice labs are not well organized as previous course. The optional labs have less hands on activities which gives less hands on activities for the learner. It would be great if there could be more activities that the learner could perform instead of just looking at the code

It's okay for intro but I feel need in more practice. Don't expect to get skills from this course. You will get a lot of new theoretical information with just a scratch of practice. So I suggest to mix this course with some good books + practice.

it was a really great theorical course but the practical parts were too small. so I know alot about the concepts and I can calculate stuff but i can not implement them that much , and I'd rather the course to have more difiicult math! well done!

Could give a bit more focus on the coding part. What I mean is rather than giving notebooks with already written code, you could teach the code to the students and have them write it themselves.

It is really interesting, I think I have miss more focus on connecting everything together from the function f_wb to cost function to deltas to preditions. The labs are not very challenging.

Code cannot be transferred easily to jupyter notebooks or google colab. The course was 90% theory. Needs more real-world practice and projects. Other than that Theory was very good.

the course is well structured and you'll gain the fundamentals of ML. But the practical labs should be more challenging in my opinion. I am looking forward to the next course.

There is very less practice content for students. I hope keeping more graded notebooks would be helpful for students to get more practice and understand the concepts better.

The course is interesting but focuses on very theoretical aspects, more suited to academia than the working world. Unfortunately, the audio is not the best.

This course requires Python programming experience using the libraries this course needs. It is not suitable for learners with basic coding experience.

Much more of the lesson content needs to be focused on the code, rather than the underlying maths, seeing as it is purely code which is assessed.

Love the trainer. Would have been nice to start coding earlier. By the time we were called to actually write code, it was way too late.