Chevron Left
Retour à Apprentissage mechanique pratique

Avis et commentaires pour l'étudiant pour Apprentissage mechanique pratique par Université Johns-Hopkins

4.5
2,667 notes
500 avis

À propos du cours

One of the most common tasks performed by data scientists and data analysts are prediction and machine learning. This course will cover the basic components of building and applying prediction functions with an emphasis on practical applications. The course will provide basic grounding in concepts such as training and tests sets, overfitting, and error rates. The course will also introduce a range of model based and algorithmic machine learning methods including regression, classification trees, Naive Bayes, and random forests. The course will cover the complete process of building prediction functions including data collection, feature creation, algorithms, and evaluation....

Meilleurs avis

JC

Jan 17, 2017

excellent course. Be prepared to learn a lot if you work hard and don't give up if you think it is hard, just continue thinking, and interact with other students and tutors + Google and Stackoverflow!

AD

Mar 01, 2017

Issues of every stage of the construction of learning machine model, as well as issues with several different machine learning methods are well and in fine yet very understandable detail explained.

Filtrer par :

426 - 450 sur 491 Examens pour Apprentissage mechanique pratique

par Robert C

Aug 01, 2017

This course needs swirl assignments. I did fine on the quizzes and assignments, but I only feel like I learned a minimal amount of machine learning, even practical machine learning.

par Rafael M

Mar 30, 2016

The course feels rushed. I understand teaching Machine Learning in 4 weeks is impossible, but then maybe the course needs to have a narrower yet deeper scope rather than throw at you many concepts without details. e.g. trees, random forests, bagging and boosting all in 10 minutes each? Impossible...

So, as opposed to creating machine learning intuition I feel the course became an R package code book.

par José A G R

Feb 05, 2017

Superfluous but the existence of the package "caret" covers the gap of other libraries like "skilearn" of python

par Baha`a A D

Oct 20, 2016

Good enough to open up mind of researcher

par Ivana L

Feb 24, 2016

Compared to previous two courses in specialization this one is far worse - it is more of excursion into used methods than actual learning using any of mentioned methods in enough detail to be able to do meaningful analyses.

par Miguel J d S P

May 19, 2017

I didn't enjoy the supporting materials and the quizzes weren't very interesting. The final project was fine.

The subject is super interesting.

par Henrique C A

Oct 14, 2016

Exercises could be more complete, and some are outdated for latest R, giving slightly different results.

par Serene S

Apr 29, 2016

too easy

par Yury Z

Feb 03, 2016

I'm somewhat disappointed. I attend almost all other courses in this specialization (except of "data product") and this one is, on my opinion, the weakest one. A lot of links to useful information though. This is more reference guide rather than a real training course.

I can say even more, initially I start other courses of this specialization just because they were marked as strong prerequisite to this one. For now, I think all other courses of the specialization were much more valuable for me than this one.

I've also took Andrew Ng course on Machine Learning in the past, and my learning experience was much better. In lectures on some concepts (like regularization) I'm pretty sure I would not understand anything if I had not been familiar with the subject before..

par Hongzhi Z

Jan 03, 2018

All the formulas and code in slides are too abstract. If can be more charts to interpret that will be better.

par Brian F

Aug 16, 2017

There was some good material in here, but it was rushed and is deserving of a much longer course - especially compared to some of the other modules in this course.

par Kyle H

May 09, 2018

A brisk introduction to some of the basics of Machine Learning. Will leave with an understanding of a few ways to use the caret package.

par BAUYRJAN J

Mar 01, 2017

Instructor rushes the course and does not explain much in the same level of details as respective quiz requires

par Eduardo P

Apr 14, 2017

This is such a cornerstone topic to the Data Science Specialization that I think it deserves a better designed and more polished curriculum. The subject is so extensive that it might be worthy to split the contents in two courses. Finally, I would like to suggest the authors of the course modeling the curriculum following the amazing treatment of the subject found in "Introduction to Statistical Learning" by Hastie, Tibshiriani et. al.

par Christoph G

Dec 04, 2016

The topic is too big, for one course from my point of view.

par Andrew W

Feb 10, 2017

The videos are really tutorials on R functions for machine learning and data wrangling. A good substitute for "Machine Learning" by Andrew Ng in terms of managing data sets and exploratory analysis.

par Christian B

Apr 30, 2017

First I want to thank very much the instructor in the online forum. He helped me a lot at the end of the course and his tutorials for gh-pages are excellent. He was also very fast in responding. Thank you.

The course did ultimately not really gave me what I was looking for. Maybe too may different facts and not enough depth. I am not sure that I can confidently say that I can build a ML model now. Technically I can, but the deeper understanding is missing. For example: When would I use which method (for example rf versus naive base), the last exercise about cross validation was not fully clear. Using the caret package is too high level for a learner. It would be better to see some more step by step examples. It was not clear to me what the expected error calculation in the last exercise was really looking at. Maybe what is missing a swirl exercises, not using caret. and then explaining how caret can simplify it. We also learned how to create a predictive model, but did not go into how the model gets updated and gets retrained, an important aspect of ML. i also do not see unsupervised learning to be covered.

par fabio a a l l

Nov 14, 2017

Poor supporting material in a course that tries to cover a lot in a very limited time.

par Léa F

Jan 09, 2018

Rather good overview. The contents could dig deeper into each subject, and it would improve the course a lot if some exercises in Swirl were added.

par Andrew W

Mar 13, 2018

Very interesting subject area, I think there is simply too much to cram into one course. Should consider spliting the subject into 2 courese or simply concentrate on only 1 or 2 main areas (e.g. cla

par Ariel S G

Jun 27, 2017

In my opinion, this course needs a few extra exercises.

par Christopher B

Mar 01, 2017

While the overview of the content seemed very reasonable both in scope and pacing, the lack of swirl exercises meant that the final project for the course was a bit jostling. Overall, I think this course still needs some development in the way of exercises to familiarize the student with the practical exercises associated with machine learning.

par Max M

Dec 12, 2017

Should have gone into more depth and included swirl lessons, like previous courses. The quizzes were very challenging though, so that helped.

par Francois v W

Dec 10, 2017

The course gives a decent overview of the model building process and covers a good spread of machine learning methodologies. I found that the videos focused too much on some basic/immaterial concepts at times and tended to gloss over the more in-depth or complicated sections. It would have helped if difficult concepts were explained with more examples. This meant that a lot of self study outside the lecture notes had to be done. The way that the final assignment had to be submitted on Github resulted in me spending 8 times longer on learning how to post my results than actually building the model - some more guidance here would have helped a lot as the process was very frustrating.

par Jorge B S

Jun 25, 2019

I have passed 5 courses of this specialization and I am not fully satisfied with this one. The course is a very brief introduction to practical machine learning, as the concepts are explained very fast and without a minimum level of detail. Then, most importantly, there are no swirl exercises, so it is quite difficult to put the acquired knowledge into practice. The other 4 courses I took, they all had swirl and that was great. Nevertheless, the course project is quite nice in order to face a real machine learning problem.