.

a

.

g

E

The Data Analysis with Python course is a decent intro into data handling, cleaning and some basic model building. I would say however that without prior Python or data analysis experience, it is probably quite a steep learning curve. I am from a science background with a bit of advanced maths knowledge, and the Model evaluation part got a little bit crazy even for me, I'll probably have to revisit it a few times. But you also can't expect to do a course like this and have a complete grasp of everything guaranteed, still have to be willing to do your own further reading/research outside of it, just like you would for uni. Some of the notebooks contained errors or the answers were spelled out which was a bit disappointing. My last project was graded immediately which was great, and I made sure to grade a couple extra to help other learners. Over all a great intro into some trickier topics, I enjoyed it.

I had to remove 1 star just because of the fact that a project is not included in this one. Yes, you do have labs but there you are forced to write code in way so that you don't encounter problems later in the notebook.

In a project or an assignment work, you have to play with variables and confusions and errors out of wonderland show up which lead to greater clarification.

The course in itself is great and undoubtedly good in functioning as a prerequisite for Machine Learning and surely I'd recommend it to anyone who asks for an opinion. The explanations are good and much easier to understand along with the visual demonstrations.

I'd advise that after learning anything during this course, look for some database online and play with it yourself(I didn't but had to regret cuz I'd forget the code again and again).

OVERALL : GO AHEAD, it's worth it

The final assignment for this course is frustrating because it uses Watson Studio instead of the learning environment we've used up to that point in the course. There is registration, learning curve, additional complexity. One of the final questions doesn't accept a file upload. One of the questions asks you to calculate a regplot which takes about 15-20 minutes to complete. Nothing tells you to expect that until I finally found a student comment in the forum. I was so frustrated at one point I was ready to abandon the course because I didn't know what was wrong and couldn't find any help. The issues with the final assignment need to be addressed, but I'm glad I decided to stay with it. The course presented a lot of good material. I recommend the course, but I hope they will address the rough edges.

Overall a good introductory course. It features several interesting aspects of pandas, numpy and matplotlib. The focus lies on using statistical methods in Python, not on explaining statistics itself, bu a qualititative (short) explanation is given for all the items shown in the course.

A few things could definetly be improved:

Some parts of the videos or their accompanying notebooks have some errors and should be checked.

The course kinda suffers from only testing he students via quiz options. While the structure ad high frequency of the quizzes is to be commended and helps students stay on topic, there should be some final assignment hich consists of actual coding tasks. Not too difficult, but with some of the methods explained each week.

This was a very good course. It assumes a lot of knowledge about statistics. I haven't done stats in 15 year so I often had to leave the platform to review statistics to better understand. Some optional statistics reviews, especially with polynomial regression and ridge regression would have been helpful. Also, I think the course missed an opportunity to check if learners really understand the analysis we are doing. Running the code is only a part of being good at data analysis. More questions and lab problems that require demonstrating an understanding of what the analysis means would make the lessons more meaningful and more likely to stick with learners.

The course like every other course in the specialization is a little too fast for me. The videos are way too short (averaging around 5 minutes) and there is way too much stuff in the videos for you to be able to absorb properly. But, the good thing is you'll still learn a lot from this course. Things got really fast mid of 4th week onwards for me, like explaining ridge regression and very complex topics without any proper introduction has left me kind of clueless. But a course should be rated according to how much you really learned and despite of all of the things above that and some more, I still learned a lot from this course so 4 stars for that.

First of all I would like to thank all the instructors for creating this course. But I feel the couse content lacks a lot of clarity if someone is taking this course for the first time. The video lectures lacks a lot of the important theoritical and coding concepts. And in the Hands-On lab there were many coding sections again if someone is doing this course for the first time it will be very hard for them to understand. In my opinion the the video lectures should have been a bit more lengthier with more explanation of the coding part as why some coding sections or some lines of codes were added and also a bit more theoretical explanations.

I completed the Data Analysis with Python course, and I must say it was an exceptional learning experience. This course provided a solid foundation in data analysis using Python and covered a wide range of topics. The course content was well-structured, and the instructors did a fantastic job of explaining complex concepts in a simple and understandable manner. I highly recommend the Data Analysis with Python course to anyone looking to enhance their data analysis skills. Whether you're a beginner or have some experience, this course offers a comprehensive and practical approach to data analysis using Python. Thanks Shubham Sanger

A good starting point.

Some of the concepts could have been explained more clearly, I have decent mathematics understanding and sometimes still felt like I was hopping from this to that (regarding the codes). I understand that they're trying to teach many things in basic level, but total video time was about only 1 hour for the whole course... I wouldn't mind watching a little more or getting additional reading materials to get the context & familiarize myself with the codes (I do additional practices on my own so that's fine, but directed-study is always nice, and easier)

The labs were really helpful though, so I'd say go for it!

I think the course was good, but the complexity level of the labs was a bit high. I mean, the leap in skill level required could have been made more easier. There are many new functions utilized in the labs that we have not been made familiar with. So, a lot of documentation-perusal and sifting of other online resources was required, especially with the Polynomials and Ridge Regression, the last Lab. I think the contents of the last Lab (Model Evaluation & Refinement) should be elaborated on and explained with greater clarity, introducing new functions and code-parts along the way.

One of the few good courses I have had. I learned a lot, used much of it in the labs. The lab for Week 5 was very confusing, as was the final one. The other labs were great, but Week 5 and the final were very disjointed and uneven. There were several things I had hoped they would put in the lab, but there was no first-to-last example lab, which is what I wanted. Without actual instructors (as in live training) this should be expected, and if I had paid for this I would be upset. Problems 7-10 on week 5 are garbage!

Still, one of the best on Coursera, from my limited point of view.

The course is great but they don't really give enough information about some stuff, I hoped they would explain what is really the goal of alot of snipets of code and which part does what in a deeper level instead of just scratching the surface,

i had to teach myself somestuff and it was a little challenging for a while specially that i don't have statistics background i, Iam not talking about more content i mean more info more details in other words what's obvious for the developers who provided this course isn't that obvious for new learners

Of course I've learnt a great deal about data analysis using python in this course. The course videos were made in a way that even the most difficult topics could one learn very easily. And after-module-labs were great to test the topics learnt.

One thing that stopped me from giving full 5-star was the final assignment. It was way too simple in my opinion. Most of the discussed topics weren't even there. I guess it would be much better to make the final assignment a little bit more thorough and to some extent, more difficult.

The course is overall very helpful to learn Data Science with Python while it does require foundations for statistics for this module, so it appears difficult to understand some mathmetical concepts for beginners. Thus I suggest some more detail explainations/practices for core parts like model development.

Moreover, there are some mistakes/typos in labs, e.g. Week5's Model Evaluation and Refinement, though most of them are minor. Also some libaries are outdated (discovered thourgh warning outcomes), which may need updating.

Good because provided breadth - teaching lots of different data analytics tools. The cons were that it didn't actually force you to code until the final product, and it also tried to do way too much in one course. I wish it just went more in depth into beginner topics like cross-tabs and linear regression, as opposed to trying to cover introductory stuff as well as beginning machine learning in one course - which caused the course to sacrifice depth - a deep understanding of any given topic.

Course is good, but in the slides there are some typos in function names or libraries. This could be a problem for some newbies to python.

In one lab deprecated parameter is given which is not used for a long time (Perform a grid search for the alpha parameter and the normalization parameter, then find the best values of the parameters. Normalization was used in version 0.18, right now is 1.3 ). As it was given in the lab as a task I have spend a lot of time to get, why it doesn't work.

I love this course! I think it's well organized. And they made sure you really learn in the lab. I'm very hands-on when it comes to embedding important skills (via the lab exercises). I do wish they would associate the terms with, say, statistics or machine-learning, so I would delve into more research -- even more than necessary. Not everyone who wants to learn Python is already well-versed with the prerequisites. But overall, the course is thorough enough and well-articulated.

This course is good if you already have some experience in Python and its structures, or if you have some knowledge in programming. You will learn some basic data manipulation and exploration techniques and also start with some of the model evaluation metrics in order to assess the (regression) models created. Overall good experience. If you already have some knowledge of Python SciKit Learn and Pandas, you could easily cram this course in 2 days (all-in) without too much sweat.

It has been a fantastic experience to have gone through this course materials. Although I found the lecture videos quite quick to the extent that we fail to understand the concept well. But while going through the labs carefully, I was able to get the concepts right. So only because the lab part was well organized, the course was helpful to me. But had it been the lectures alone, then it would have been difficult to grasp all the concepts clearly.