May 10, 2020
The course had helped in understanding the concepts of NumPy and pandas. The assignments were so helpful to apply these concepts which provide an in-depth understanding of the Numpy as well as pandans
Mar 16, 2018
overall the good introductory course of python for data science but i feel it should have covered the basics in more details .specially for the ones who do not have any prior programming background .
By Saadman S•
Sep 16, 2020
Statistical stuffs are really tough, it's hard to understand without any background also the assignment materials should be discussed more, they should be included in the course.
By Colleen K•
Sep 22, 2018
I learned a lot by doing assignments, but the course materials are not helpful. Stackflow and Python documents guide me much more than the course itself.
By Khairul A•
Aug 10, 2020
Too fast explanation
By Erico L•
Mar 02, 2019
I don't think I've learned much along the course. I had to pick a few concepts here and there, but I don't think that the way in which those are explained would stick.
Also, the course seems rushed: I'm not sure what the end game of these courses is, but I think it's an incredible wasted opportunity when it comes to MOOCs, as there could be more lengthy videos and more and better ungraded exercises (something that in this particular course do not exist) and much, much better explained assignments (I guess adding there the info from the forums by the teaching stuff would not hurt).
For being a course of intermediate level, the videos and explanations are too short; there are even places where things are left totally unexplained.
Even if it's supposed (and even encouraged) that the students seek information on their own, the lack of context in some places makes it rather difficult. this is specialy more so with the questions that are interwined in the videos, as normally in order to answer them corretly you have to go out and find the related info (something that totally disrupts watching the videos).
finally, the assignments are a wreckage; some of the questions are incredible difficult to understand, if not out right impossible. The fact that there's a lot of information added to the forums by the etaching stuff, up to the point that the more complicated questions are easily answered with that same infromation, proves this.
I do think there are examples of courses in Coursera: I recently completed "Mathematics for Machine Learning: Linear Algebra" and even thought I don't think it's not without its issues, I find it a much more challenging, entertaining and fun course, that covers in a good way its subject.
I have to commend the people from the teaching stuff that are in the forums, thought, as it's the only course in which I found people from the teaching area activelly participating, and helping the students.
By Jun-Hoe L•
Oct 09, 2020
Decided to rate this course after I've gone through all 5 courses in the Speclisation. I originally completed this course in January 2020.
So from someone who has completed this Specialisation, I'd say this 5 courses are not worth it.
Here's how I would rank the courses from best to worst:
1. Social Network Analysis: 4.5 stars
2. Applied Machine Learning: 3.5-4 stars
3. Applied Text Mining: 3.5 stars
4. Intro to Data Science: 2
Note that that worst courses are those handled by Professor Brooks himself. His video lectures tend to very superficial (or once in a while, unnecessarily detailed like going into the backend of matplotlib). The assignments on the other hand, are somewhat challenging and go way beyond the video lectures. And that's why you see many comments asking what's the point of purchasing this course when you spend 95% of the time googling? Which is made worse by the outdated autograder which uses and old panda version, and makes googling harder since you had to revert to outdated code.
My advice: Unless you really want the Specialisation cert, I think you should look elsewhere to learn pandas.
By Zayd A•
May 28, 2019
I had done "Python for Everybody" from Charles Severance which I had found excellent, with the instructor being passionate and the pace being just about right. I had assumed it would be similar for "Introduction to Data Science in Python", but that wasn't case. The delivery of the course is at a very very fast pace, you don't even have time to stop and absorb the functions and methods that you are supposed to learn. The instructor and the research assistant will list the functions and methods one after the other without pausing. The assignment is then extremely hard with no resemblance to the material in the course (I couldn't do it even after having reviewed the videos). After holding on for the first 2 weeks (it's a very useful topic after all), I gave up and decided to learn from the "learning the Pandas library book", which is a very good summary of the main Pandas functions and methods (and which was recommended by Dr Christopher Brooks), and I was able to follow it very easily.
By Lucas C•
Sep 01, 2019
Overall: I felt this course was useful but pretty time-consuming. The course had relatively limited taught material and relied a lot on searching & self-studying. If you have a fair amount of time it is a good choice.
Pros: You learn through doing assignments which are well supported by mentors/community. Also, you get used to studying through googling problems and learning from websites such as Stackoverflow.
Cons: Whilst this learning method definitely had its merits, it could be quite time-consuming for someone seeking to gain introductory-level skills quickly. You could find yourself in situations where you spend hours searching for something quite elementary and could easily have been taught to you, which could be frustrating. I personally think this course could be improved by adding a bit more small quizzes for beginners to play around with the basics, before requiring them to self-learn through searches.
By Markus Z•
Aug 07, 2020
Compared to the previous course I have taken at University of Michigan the content was ok but how it was taught I didn't like. Just reading rapidly the text of the Jupyter Notebook is not enough from my point of view. Ok you can find out the stuff yourself but why take then this course and don't go directly to stack overflow.
You just get weird replies from the auto grader and search through the forum to get any idea why you didn't pass. And if you pass, you will never know if your solution was the proper way to solve this task....
By Daniel D•
Jan 29, 2020
I agree with some reviews saying that course was mostly limited to self-learning. Videos were rushed and learning mostly limited to self-studying. Assignments descriptions were confusing and not well explained, not to mention that it takes hours to figure out why correct solution is not accepted. I'd say writing code (correctly) takes 4 hours but then you need 8 hours to figure out why your answer is not accepted.
By Carl M•
Nov 14, 2019
Poorly worded questions (that are mentioned throughout the discussion board), older version of pandas and the course resources don't help you with course. Get ready to 'learn' by looking in StackOverflow or reading the volumes upon volumes of python/pandas documentation. In other words, expect to spend 15 hours a week per week (obviously it will vary)
By Brent D•
Aug 12, 2020
Lectures do not reflect what is required to complete the assignments. Much of the learning is left to independent study by the student. Assignment questions are too vague and frequently require parsing through class discussions to determine the answer the auto-grader is looking for.
By Olena K•
Mar 28, 2019
The lectures are not good. They go too quickly. They're about 5 minutes long, but you have to stop every minute or 30 seconds and rewind to understand what the instructor is saying. He just goes way too fast, and it's very frustrating. Really ruins the experience.
By Yizi Z•
Nov 09, 2018
There is only few minutes taught video courses each week, although the reading materials and topics are quite interesting. The learning of python coding rely heavily on your own trial and error, which you could do even without this course.
By Saurabh C•
Sep 03, 2020
The level of course content and assignments is not at all similar. The course contents need to be revised, seems like the professor assumes we know everything about the topic. Also the teaching speed is extremely fast. Very Disappointed!
By BOORLA V N•
Sep 15, 2020
The instructor seems like he's reading out points of a book. No proper explanation of tools used followed by assignments so hard compared to what being taught in classes
By Chris L•
Dec 17, 2019
It never felt like the material was covered in enough depth to give me confidence in the ability to do the assignments.
By Lee S•
Dec 19, 2018
Starts off well, then escalates way too quickly. Assignment 4 is incredibly complex and has poor guidance notes.
By Avneesh D•
Aug 07, 2020
The assignments were way more complicated than the examples used during the lectures.
By Georgios A•
Jan 07, 2019
Too difficult, poor connection between lectures and assignments
By Divyanshu P•
Sep 13, 2020
Insanely fast paced course.. Needs Improvement
By Kannan S•
Nov 21, 2016
This is in fact the worst course so far. Mainly because of auto grader. Here are my reasons.
Actually I did not complete the course at all. But I suddenly got a message saying that I have completed the course. I was working on the first problem of the 4th assignment. I did a provisional submission to see if my answer was right. Auto grader reported the grade for the 3rd assignment and said that I have passed the course. Any submission I did after that was not graded at all.
The assignments are not very clear. Looks like I had a older version of the questions while others had a different version. I was stuck in a particular problem because auto grader did not give me a clear feedback as to why I was incorrect. I wasted too much time on this already.
The assignments require too much research outside what is covered in the videos. I don't feel that is right. The assignment requires that we research on Stack Overflow and Pandas documentation. I strongly feel that such activities should be performed only outside the course work when we try to solve real world problems. Course assignments should be reasonably given based only the materials covered in video. This was taking too much time.
The discussion forums are not giving clear hints. When we are stuck in a problem, we are not able to proceed further. I still son't know the answers for certain problems because the coordinators do not explain the answers well. When we complete assignments we don't get to see the instructor's solution.
The video instructions were too fast paced. The instructors do not pause and explain critical aspects of the code.
Overall I am very disappointed with this course. There are much better videos on Youtube and Lynda than this . I am sorry. I never thought it would be this bad. The first course on Python from University of Michigan was really very good.
By Joseph G•
Mar 03, 2018
Not sure whether this course is trying to reach data science or Python, but it does a poor job at both.
The class is a light-speed tour through NumPy and Pandas, definitely not for the neophyte Python developer (which I am not). There's 30-40 mins of lecture each week that's basically lightly narrated typing into a Jupyter notebook with only the slightest bit of additional explanation about what the instructor is doing, although the material covered is substantial. There's lot of important details that are glossed over -- forcing the student to pause the lecture and do offline research to understand what just happened.
Similarly, the assignments address and cover beyond the material covered, but the instruction is scarcely sufficient to understand the concepts required to complete them, so lots of Stack Overview and other research is required. And the automated grader, as expected, is completely literal so for complex problems, not much help in validating whether you're on the right track. Assignments take many multiples of the estimated time.
And because even for paying students (such as myself), you never get access to an answer key even after the assignment is due, you have no idea how closely your solution conformed to best practices, even if you arrived at the right answer. For coding, this makes all of the difference, particularly with large datasets that could consume considerable computing resources if not done correctly. I'm told this is because of potential cheating by learners.
How would I change this course? Simple: 3x more lecture material to actually explain what's going on, or down-scope the class so that the existing lecture time becomes adequate for the material.
By Hari B•
Apr 09, 2017
Very poor course, badly taught and terrible value for money. The lessons are brief beyond any form of reasonableness, the teacher seems completely unconnected with his students. There is no detail at all and no logical progression. I took and passed this course with a view to doing the specialisation but I'm not going to waste any more money on University of Michigan courses. I've found similar courses on other platforms which cover the same material. The assignments were awful, in some cases they covered material to be presented the following week, in others the questions were wrongly stated and did not match the output from the machine grading. The machine grading itself gave you no clue as to where you went wrong. I'm not talking about the odd question here or there, I'm talking about consistently throughout every assignment. I don't normally, in fact ever, leave bad reviews, I usually just chalk it up to experience and move on but in this case, the course was so bad, I had to say something. I've done two other courses on Coursera with Rice University and the difference to this course is huge, while I would wholeheartedly recommend the Rice Intro to Python courses, Don't do this course, it is not coherently presented or graded. The mentors in the forum tried their best but even they had to admit the grading system was riddled with errors. Absolute rubbish, avoid and spend your money elsewhere.
By Albi K•
Oct 30, 2019
I have just completed this course. I have learned quite a bit about the pandas library and that has nothing to do with this course.
The lectures seemed to be scripted; and extremely condensed. At best, they can be used as a sparse reference manual for some undefined subset of the pandas library.
The assignment 4 instructions encourage googling things. Basically "go forth and figure it out on your own" ... why would I need a full course for that piece of advice?
The autograder seems to forbid the usage of certain lines of code in Assignment 4. It will reject your answer and give you no feedback whatsoever with respect to the reasons why your answer was rejected.
As well, it has inconsistencies that will cost you time. The question on the recession_start() function will be graded as correct if recession_start() outputs a certain value, say x. Yet, in another question recession_start() is expected to output some other value y. Go figure. Not even a warning about it.
So, to sum up the salient points:
1. Autograder has holes.
2.Extremely condensed scripted lectures and sparsely sprinkled with practical advice.
3. Useful for letting you know that pandas exist.
By Vikram A•
Aug 08, 2017
This course is poorly done, and I'm sorry but in no way close to an intermediate level. Even knowing a fair amount of python, I struggled with learning from this course. I find it ironic that the teacher specializes in education and mostly sits in a chair and speaks code at you. There are very few visual aids to help.
Furthermore, individual topics are not broken down well, showing you how to develop a mastery over the fundamental data objects like a data frame before moving on to the next. Code that is demonstrated is typed out unreasonably fast, and very few examples are done on how to properly access the elements in different ways. The video where the grad student/post doc spits out code 3 lines a minutes made me laugh at how ridiculous it was as if it were an explanation.
I ended up very frustrated with this course, and I'm not convinced it's all me or my inability to learn. I suggest learning data science in python from another site, I'm already finding a different class much better and more understandable. Your mileage will obviously vary.