.

Really good conceptual teaching of ANNs and decision trees, but it's a little lacking in the Python implementation. It teaches you how to program an ANN with any number of layers/neurons, but there is no mention of finding the "optimal" number of each. The last week on decision trees and ensemble models feels rushed as there is only one lab and required assignment, so it completely misses the Python implementation of XGBoost. However, it teaches the essential functions in each library, so one can easily continue his or her learning with Kaggle competitions and Stack Overflow. In the end, it's meant to introduce working professionals to the most common ML models in the world today, and it does that very well, but not much more.

This course is brilliance personified especially the intuition (which is the primary focus). The reason for a 4-star rating is, that it ended quickly, and it does not cover the codes in detail, but rather the logic on 'why we do what we do'. Andrew Ng drips knowledge and passion . I only wish he formulates a course named "AI-Scientist" with a one year completion time, that covers topics right from basics of Python, basic math, advanced math, ML, DL, NLP, MLOPs through and through. I am excited to jump into my next course :) Thank you Andrew Ng :)

I am missing handouts from the course and also access to the labs upon completion of the course. It was great that practice labs were offering a lot of help for the student, but I am afraid that too much material was given and the assignment was only about typing the given equation. It leaves me with the feeling that it would be difficult to apply the knowledge from the course to the real word problem, especially that I do not have any code available after the course is finished...

Excellent explanation of the concepts by Andrew Ng. However, like other reviewers, I find the last week a little bit rushed and, as compared to the first course of the specialization, this course feels a little...lacking, not in the sense of the information being taught but how the information is being presented (eg. the effort put into making quizzes and labs is subpar ). note: subpar of best is still good so I'd still recommend this one to anyone.

It was a very good and interesting course. I learned a lot about machine learning algorithms.

Compared to the first course "Supervised Machine Learning: Regression and Classification" there were a few things missing:

1- Practical exercises

2- Quizzes during the videos

Although you can learn the theoretical content very well in this course, in my opinion there is a lack of opportunities to practically apply and practice the knowledge you have learned.

Explanation: 5 starts Assignments: 2.5 or 3 stars

Thus, overall 4 stars. Andrew did an excellent job in explaining the concepts. However, the assignments, in my opinion, were

too easy (almost just running the cells or typing what was shown in lecture videos). I believe challenging

assignments are an important aspect of any course which this course lacks (unfortunately).

The course material and instructor were very good. I just have one complaint for this course... The quizzes are too easy but sure they capture whether you have paid attention in lectures or not, so that is fine. Also a suggestion - If we could be provided with the lecture ppts it would be really helpful revising.

One of the best mahine learning courses but some of the parts were boring in the middle like week 3 lectures and assignments where you just had theory classes. Some of the parts were very great like all the Decision Tree classes were too good i will never forget how decision tree works now.

Thank you Sir Andrew

Some of the topics were not explained clearly and found them quite complicated. Topics:

1. XGBoost

2. Bias and Variance

Also, I did not understand why most of assignments were about writing codes for functions that are already implemented in open source libraries and packages!

Great course. Really easy to follow and it's a good starting point for learning about ML, assuming you already know linear and logic regression. I wish there were more programming assignments and more lectures on TensorFlow though.

I wish there was more on the practical use and coding part. And the labs could be a bit more explained. Apart from all that, its an amazing course and it has helped me understand many concepts clearly from the ground up.

Truly peaking, I used hints very often because I was not familiar with the Phyton. A optional week as a zero week should be introduce to teach basic python for the students that have no previous interaction with python.

I liked but maybe the level is too basic. Anyway for those who have never seen machine learning, it is a beautiful gentle introduction to all the main concepts explained in a super clear and simple manner.

Course material is good and flows well, but are ANNs and decision trees the only advanced algos? Loved the parts on model bias/variance determination and how to fix the model based on the determination.

Every concept was explained brilliantly by Andrew Sir. Though the lab sessions code is much simpler so I feel the code level should be increased. Overall the best course to start your ML journey with :)

Beginner level course, explained simple concepts on neural network specifically Multi-Layer Perceptron & Decision Tree, nothing advanced topics covered. But the explanation is very very good.

It's a great course, really phenomenal but it lacks a little on the implementation side. It would be better if it had 1 or 2 medium-level hands-on projects included in its structure.

I felt the lab assignment in the Decision trees section was a little too fast to comprehend. Otherwise, it was an excellent course with just the necessary theory and intuition.

Good to review some concepts if you have an Intermediate level of knowledge and excelent if you're new to the area. Still very light on mathematics, but the course is excelent

Very good didactical approach. The labs are straight-forward but test programming skills more than AI expertise. Editing and error checking in the notebooks is poor.

Very good learning experience. Bit difficult than the previous course on supervised learning, but concepts can be understood if we listen and understand carefully.

very good course but it needs lab practice , it looks like we just run the code , aand if we want to write by our self it is hard . but thank you so much

It would be much better if you link a relevant applied AI course with this package. The course is more of theory, than practical application.