Chevron Left
返回到 Applied Machine Learning in Python

学生对 密歇根大学 提供的 Applied Machine Learning in Python 的评价和反馈

4.6
6,161 个评分
1,108 条评论

课程概述

This course will introduce the learner to applied machine learning, focusing more on the techniques and methods than on the statistics behind these methods. The course will start with a discussion of how machine learning is different than descriptive statistics, and introduce the scikit learn toolkit through a tutorial. The issue of dimensionality of data will be discussed, and the task of clustering data, as well as evaluating those clusters, will be tackled. Supervised approaches for creating predictive models will be described, and learners will be able to apply the scikit learn predictive modelling methods while understanding process issues related to data generalizability (e.g. cross validation, overfitting). The course will end with a look at more advanced techniques, such as building ensembles, and practical limitations of predictive models. By the end of this course, students will be able to identify the difference between a supervised (classification) and unsupervised (clustering) technique, identify which technique they need to apply for a particular dataset and need, engineer features to meet that need, and write python code to carry out an analysis. This course should be taken after Introduction to Data Science in Python and Applied Plotting, Charting & Data Representation in Python and before Applied Text Mining in Python and Applied Social Analysis in Python....

热门审阅

FL

Oct 14, 2017

Very well structured course, and very interesting too! Has made me want to pursue a career in machine learning. I originally just wanted to learn to program, without true goal, now I have one thanks!!

OA

Sep 09, 2017

This course is ideally designed for understanding, which tools you can use to do machine learning tasks in python. However, for deep understanding ML algorithms you should take more math based courses

筛选依据:

801 - Applied Machine Learning in Python 的 825 个评论(共 1,092 个)

创建者 parmar p

May 18, 2020

nice

创建者 Miriam Y R L

Dec 27, 2019

good

创建者 Light0617

May 13, 2019

nice

创建者 Shishir N

Jan 09, 2019

N

i

c

e

创建者 Jimut B P

Oct 08, 2018

Nice

创建者 Yi-Yang L

Jul 03, 2017

Nice

创建者 SURAJ K

Jun 23, 2020

osm

创建者 Shilpi G

Jun 02, 2019

...

创建者 Magdiel B d N A

May 10, 2019

ok

创建者 PREDEEP K

Nov 24, 2018

ok

创建者 Andrew G

May 16, 2019

T

创建者 Junaid L S

May 14, 2019

G

创建者 Thomas W

Mar 06, 2018

g

创建者 Oleh Z

Feb 27, 2018

G

创建者 Piotr B

Jun 01, 2017

a

创建者 Jun-Hoe L

Jun 03, 2020

My actual rating is 3.5 stars. This is the best course yet in this Specialization.

Pros: I prefer Professor Collin-Thompson's delivery compared to Professor Brook in the previous modules. I think he gives a good overview and sufficient depth for an applied course, compared to Professor Brooks which I find to be quite superficial most of the time, and weirdly detailed in other parts. Assignment is good enough for reinforcement learning and definitely better planned. I also appreciate the link to additional readings which are quite informative.

Cons: Assignment auto-grader. This is still the biggest letdown of all the courses in this specialization Codes which work on your laptop or suggested elsewhere on Stackoverflow etc fails to pass the autograder, so 30-40% of the time of the assignment is spent on wrangling the code to pass the autograder.

Note: If i haven't taken a Machine Learning course by Professor Andrew Ng, this course would definitely be much harder. This course doesn't go to much into the background knowledge,and they mentioned this many times. But I appreciated the applied aspect, since this was what I was looking for.

创建者 Carolyn O

Jan 19, 2020

I had no ML background, although I have the math the models are based on. The material seemed more than week's worth for a couple of weeks. The quizzes make sure you don't miss the key points you need to take away and need for the assignment. Most information or key words are in the slides, but course expects you to be independent enough (intermediate) to learn closely related ideas on your own via StackOverFlow and discussion forums. The discussion forums were especially helpful for this course, but then online discussions makes it more studying alone. Discussions helped me trouble-shoot and get better ideas how to approach the problems generally. I can explore and use ML and sklearn on my own, which thankfully seems to be a goal of this professor. No material could be left out, but when more videos, better longer time estimate for the week would be nice.

创建者 YYuan

Nov 28, 2019

This course involves lots of concepts and algorithms in machine learning. As it is said by the teacher, for time, effort and aim limitations, this course only involves basic concepts and usage of sci-kit learn. It is a good hand-on course for beginners. Assignments are not so challenging compared with the previous two courses in the same specialization. I just finish assignments by following the module code in the course. I feel like not study as much as I expected through the assignment. I hope assignments can be changed by varieties and difficulties to let students know how a machine learning project is like and how the evaluation works but not simply call the precision/accuracy/recall function and the assignment finishes. Generally, you still learn a lot if you want 'applied machine learning'

创建者 Guo X W

Jun 12, 2020

I personally enjoyed this course much more than the previous 2 courses in the specialisation. Overall, this course is ambitious and covers a lot of different algorithms. For each algorithm, a brief intuition is provided and we are taught how to code in Python. For this course, I felt that the assignments were a closer fit to the content covered in the videos (unlike the previous courses where the assignments required much more independent learning). However, this course will not provide the mathematical rigour that some learners may expect. Furthermore, the amount of content covered could be a bit overwhelming. It would be useful if the instructor could summarise the different steps we should take when faced with a ML problem, esp. for deciding which algorithm to use (since so many were covered)

创建者 Zuha A

Sep 01, 2018

if you have a conceptual knowledge about Machine Learning algorithms, or at least supervise learning, this course would be very helpful for you. Otherwise, you are wasting your time.

This course is a programming session , helping you to implement the complicated machine learning algorithms using simple tools, without diving in any details or explain any mathematical backgrounds. So you supposed to build these fundamentals before coming here. For me, I took the wonderful course of Andrew Ng before this.

Furthermore, the course is very structured and organized, and its material, quizzes and assignments are greet , thus I consider their notebooks such a good reference I'll back to it every time I solve a ML problem.

创建者 Nicolás C

Jan 15, 2018

Very good course. The content is excellent. You can get a good understanding of many popular Machine Learning algorithms. Maybe the most valuable concept you can learn is how to evaluate a classification model. It is also an applied course, so anyone more concerned of the applications than the theory will enjoy it.

The only drawback is that the evaluation of the assigments is done automatically, and you can have frustrating limitations for an answer that is correct but that is not EXACTLY as expected (I mean even the data structer have to much perfectly). The server also have quite restrictive memory limitations and the error messages are not always very helpful, but the staff will help you if you insist enough.

创建者 Marcel P

Jan 28, 2019

The course is an excellent tour of machine learning methods. The best thing is that it provides the python codes for various applications of machine learning. These can represent a great starting point for real applications. The significant parameters of each model are explained and the usage of the main models is well depicted. However, the course is very dense and I think it should have been divided in 6-8 weeks. At least the unsupervised learning part, which is optional in the Week 4 should have a dedicated week, with assignments. Before doing this course I recommend something like the course of Andrew Ng (without that one, for me it would have been more difficult to follow this one).

创建者 Ian R

Nov 11, 2019

I found the course to be a little bit too much of a whirlwind for me to get much more than the broadest strokes out of it. A lot of the topics covered were mentioned very briefly without much explanation of when or how they should be applied - especially week three felt like a barrage of "this exists, this exists, this exists..." without much explanation, and I don't think I'll retain very much of it. The Week 4 assignment, however, was adequately challenging and did give me cause to go back, review and dig deeper into many of the topics covered previously.

创建者 Paulo C

May 03, 2020

Overall a good course! It was really what I was looking for: main focus is on how to apply algorithms and pros and cons of each model, instead of exhaustively explaining the theory behind each one, like some others courses do. The downside was the grade system. The platform has a lot of potential, but crashes all the time and there are many errors to troubleshoot when submitting assignments. The time invested to troubleshoot these problems was really frustrating, and probably the main reason i won't continue with the specialization.