Avis et commentaires pour d'étudiants pour Apprentissage mechanique pratique par Université Johns-Hopkins

Does not give much intuition around the subject. I found the lectures a bit uninspiring. Lots of powerpoint (just text, no images or visuals really) and the lecturer just underlined the words he was talking about as he read the powerpoint out. I found the Udacity Intro to Machine Learning course gave a much better intuition and understanding of this subject. We also had slides on how to split data into a training and testing set on pretty much EVERY lecture - what a waste of time!

I was rather disappointed with this course. I guess it fills the objective of getting you using the caret package and getting you started with some examples. However to understand what you are doing you should defintively go somewhere else. I definitively missed some swirl exercises and more flow diagrams in the slides. It felt for me as I was just copypasting some code from the slides. The course does clearly give some good literature and places to go for details.

This course is out of date. The videos and books are better than many Coursera classes, but I found the quizzes and projects to be a giant time suck because required packages need certain versions of R or are no longer maintained or some other nonsense. I was spending twice as much time trying to get the data and packages to work in R Studio than I was actually coding or thinking about what the quiz question was asking.

This course is getting too old. Some assignments are impossible to do since modern implementation of packages used are getting a COMPLETELY different answer. The theory is ok, if a bit all over the place, but it's extremely frustrating believing you did something wrong just cause your answers are better than the answers the quizes believe they should be.

Absolutely useless random un-explained list of facts and advices that is thrown to a learner without any attempt to give a systematic approach. Pure waste of time and effort. Can only be suitable to those, who already know the subject well and can use some additional facts that are randomly presented in this "course".

Personally, I found this course as the worst one among the DS Specialization courses.

Outdated.

Project description versus requirements were terrible, not sure if the new Coursera format played a role in the issues or not. Quite a few of the homework items require guessing as the answers don't align to the results of the latest tools they have you use. If the first class or three in the series was like this I wouldn't have taken the courses.

This course was a big let-down compared to other courses in the specialization. It doesn't seem like a lot of effort went into course planning and creation. Much of the content is unclear and there is little depth. course textbook, and some swirl exercises would have helped.

Excellent and useful course.

Some of the materials covered in Week 4 should be distributed to earlier week(s). The current Week 4 video coverage, quizzes, and the course project on accelerometer data is too much for the week, esp. if the student has lookup and review some key concepts from the resource links in the video slides. Video lectures are informative and easy to follow, although somewhat rushed in Week 4.

The course gives a clear explanation of why machine learning, with a goal of prediction, is different from regression. The use of the caret package in R is emphasized. Caret provides a uniform interface to many different machine learning algorithms, leaving no excuse for practitioners not to test a variety of approaches to confirm the robustness of their conclusions.

This is my favourite course in the data science. Prior to taking up this course, I have been using technical analysis to achieve my investment goal. I know how to design trading system to trade. Now with machine learning, I learned something new. System trading is reactive and machine learning is predictive. This subject is the reason why I sign up for data science.

My name is Jose Antonio. I am looking for a new Data Scientist career ( https://www.linkedin.com/in/joseantonio11)

I did this course to get new knowledge about Data Science and better understand the technology and your practical applications.

The course was excellent and the classes well taught by the Teachers.

Congratulations to Coursera team and Teachers.

This course really demystified machine learning, and provided practical steps and guidance on how to create predictive models. While I do wish there were more resources on how to tune models and investigate specific model parameters, I understand that there just wasn't enough time. I couldn't imagine a better course for a solid foundation in this skill.

Utilizing the data available to predict the information in any topic (eg: Weather forecasting, component life etc) with good accuracy is important to save lot of time and efforts. In this course, R programming has been used to predict the information through ML module. The same can be implemented in my activities which can save lot of time and efforts.

Great course, but it may take you more than the allotted 4 weeks if you intend to dig a bit deeper and pursue some of the additional resources referenced throughout the course. I would definitely recommend doing that, as there is A LOT of material to cover if you, like me, just have to know the details of what's happening behind the scenes.

This course gave a great basic understanding of some different machine learning algorithms and what they do. I now have a great practical understanding of how to implement them, and enough understanding of theory to know what I'm talking about and to be able to learn more about them in the future.

This is a very useful course in Machine Learning that teaches us how to use the R based packages such as CARET for applying machine learning techniques. The course project helps understand how these techniques are applied in real world applications and develop useful insights.

PML is a deep subject and this course is an excellent foundation for further studies. Prof Leek has taught brilliantly on the basic concepts of PML given the short time of 4 weeks. You need college level statistics to fully appreciate the theories of the PML's lectures.

It is a well designed course, for academician as well as practitioners. Syllabus of the course, covers a lot of algorithms. Course content, presentation, assignments are very practical and give a lot of knowledge, understanding and practical tips..

All the major machine learning algorithms and techniques are provided in a way that you can begin using them right away. The course project also provides an opportunity to apply the different techniques learnt in class to a rather messy dataset.

It's good that they teach you basics of machine learing in R (caret package), but it's very introductory course. I definetly recommend this course to beginner, but I also recommend taking more courses on this topic (Andrew Ng's for example).

good course, but one who is serious about data science should view this course as a starting point since machine learning is a semester long course so I'd recommend follow up with machine learning course taught from Andrew Ng out of Stanford

Really good class. I think there were some small issues with the class project. Like all real world problems it was not entirely well specified and the data was a bit odd to use for a prediction exercise because it was time series data.

This was definitely a challenging course. I learned a lot about building and testing prediction algorithms. The course also helped me overcome the feeling of intimidation by providing excellent examples and a hands-on final assignment.