Chevron Left
Retour à How to Win a Data Science Competition: Learn from Top Kagglers

Avis et commentaires pour d'étudiants pour How to Win a Data Science Competition: Learn from Top Kagglers par Université HSE

1,113 évaluations
274 avis

À propos du cours

If you want to break into competitive data science, then this course is for you! Participating in predictive modelling competitions can help you gain practical experience, improve and harness your data modelling skills in various domains such as credit, insurance, marketing, natural language processing, sales’ forecasting and computer vision to name a few. At the same time you get to do it in a competitive context against thousands of participants where each one tries to build the most predictive algorithm. Pushing each other to the limit can result in better performance and smaller prediction errors. Being able to achieve high ranks consistently can help you accelerate your career in data science. In this course, you will learn to analyse and solve competitively such predictive modelling tasks. When you finish this class, you will: - Understand how to solve predictive modelling competitions efficiently and learn which of the skills obtained can be applicable to real-world tasks. - Learn how to preprocess the data and generate new features from various sources such as text and images. - Be taught advanced feature engineering techniques like generating mean-encodings, using aggregated statistical measures or finding nearest neighbors as a means to improve your predictions. - Be able to form reliable cross validation methodologies that help you benchmark your solutions and avoid overfitting or underfitting when tested with unobserved (test) data. - Gain experience of analysing and interpreting the data. You will become aware of inconsistencies, high noise levels, errors and other data-related issues such as leakages and you will learn how to overcome them. - Acquire knowledge of different algorithms and learn how to efficiently tune their hyperparameters and achieve top performance. - Master the art of combining different machine learning models and learn how to ensemble. - Get exposed to past (winning) solutions and codes and learn how to read them. Disclaimer : This is not a machine learning online course in the general sense. This course will teach you how to get high-rank solutions against thousands of competitors with focus on practical usage of machine learning methods rather than the theoretical underpinnings behind them. Prerequisites: - Python: work with DataFrames in pandas, plot figures in matplotlib, import and train models from scikit-learn, XGBoost, LightGBM. - Machine Learning: basic understanding of linear models, K-NN, random forest, gradient boosting and neural networks. Do you have technical problems? Write to us:

Meilleurs avis

28 mars 2018

Top Kagglers gently introduce one to Data Science Competitions. One will have a great chance to learn various tips and tricks and apply them in practice throughout the course. Highly recommended!

18 févr. 2019

Really excellent. Very practical advice from top competitors. This specialization is much more information-dense than most machine learning MOOCs. You really get your money's worth.

Filtrer par :

51 - 75 sur 273 Avis pour How to Win a Data Science Competition: Learn from Top Kagglers

par Refik E

25 janv. 2018

Content is interesting but language is very difficult to understand and due to that fact the course was not engaging for me.

par Temiloluwa A

4 oct. 2019

Couldn't follow because the English Accent was difficult to understand.

par Manuel M B

22 mars 2020

Si bien los instructores tienen mucha experiencia en Kaggle, no me resultó muy util a la hora de aprender Data science para un entorno empresarial y por proyectos. Si tu objetivo es dedicarte a estas competencias te lo recomiendo, pero como parte de una especialización en ciencia de datos, No.

par Maciej

10 janv. 2019

Very dry presentation. Video is not a good medium for this material.

par Molin D

14 janv. 2021

The course itself is very good. However I have some opinion on program assignment scoring system:

1. Machine learning hosting enviroment by Coursera for jupyter notebook is SLOW.

For some task, I could download database from public source, run locally with my GPU.

For some task, I could not move whole resource because of private database, etc. Either I don't want to use Google colab (may forget to cancel GCP service charing...)

2. Classmate review suspended,

I join the this cours late, so for a long time after I submited the assignment no one to review.

better change such task to auto scoring system.

par stephane d

3 sept. 2021

A great course that covers different interesting aspects of "Feature Engineering" such as "Target encoding", "Lag features",... On the modeling side, it covers some very good algorithms such as XGBoost, LightGBM, CatBoost but I recommend to the people who follow the course not to stop on black boxes but to explore more in details by programming directly in Python some algorithms such as Decision Trees,... in order to better understand the hyperparameters to use and how to use them. I recommend this course because competition is a subject that is not often treated.

par Andrii Y

23 août 2021

Great course! Not only about competitions but on advanced machine learning in general: feature engineering, preprocessing, validation, metric optimization, hyperparameter tuning. The final project is really good, it allows to apply all the course knowledge and receive some practical data science experience. One of the best courses on Coursera I took so far!

par Kaushik P

22 déc. 2018

This course is just what I was looking for as I am really interested in competitive Machine Learning and data science. Hopefully , I will be able to perform better in competitions from now on.

But the only down side I can think of is that the programming assignments are pretty difficult at times, but none the less it was a great experience.

par Sixing H

10 déc. 2019

A very needed course in not just Kaggle competition but also machine learning. Even not for the Kaggle tips, the machine learning alone should be reason enough for taking this course. The code exercise provides excellent framework for further application in my own projects. I hope there are more such courses in coursera.

par Kirill L

20 nov. 2017

Even though, it revolves around Kaggle competitions which are usually simpler than real-life, this course is full of down-to-earth practical techniques and examples which is really valuable for me.

Idea to organize Kaggle competition as a course project is very good.

Lectors are easy to follow and nice to listen to.

par Vratislav H

29 mars 2019

It is diffifult but when you reach the end, you are glad that you were able to finish it. Because I gained a lot of knowledge and best practices. There is a lot of work but it helps you sharpen your brain. I recommend to work simultaneously on project because otherwise it will be difficult to finish it..

par YUYU L

2 sept. 2019

Teaching the clear work flow with data science project and learned some trick method at feature engineering. In the final, playing kaggle competition was funny. But, you should take the competition as early as possible, if waited week 5 to join the competition it would be very hard.

par Mark P

9 oct. 2018

This is a fantastic course for anyone looking to extend their skills in data science. Its packed full of tips and tricks and techniques that are well explained and very useful for data science. I would go so far as saying that it has been my favorite data science course OF ALL TIME!

par Lukas K

3 août 2019

One of the best course I had on Coursera so far. Really good explanation of problems you can face in DataScience competition and ability to see many useful approaches for solving the problems. The final assignment is a little bit longer, but totally worth the time.

par Francois G

11 oct. 2020

It seems to me a very good and actually quite demanding crash course in applied machine learning. It is full of very pragmatic techniques to approach data modeling with the clear goal to improve scores (by any legit means) in competitions. Well put together.

par Igor B

27 janv. 2019

This course requires much time, but gives hardcore experience in practical data science and machine learning. The final project, which is a proving ground for the acquired skills, is both an interesting competition to participate in and a real-world-task.

par Arnaud R

17 mars 2018

This course is a gold mine of knowledge and tricks for anyone working with the data science toolkit. It requires good prior knowledge of the different algorithm used and Python fluency. The course is demanding but you will get out of it so much stronger.

par Diego T B

7 nov. 2019

This is an awesome course! I really learned a lot from this top kagglers. I just have one recommendation. I think some sessions were very though and difficult to catch: the data leakage part and the Kappa metric. Try to make this even much easier.

par Yotam S

26 oct. 2019

Amazing course. Teaches the theoretical aspects of ML in within a practical point of view. Enables use to improve your models by understanding the framework much better. Not recommended as first ML course, but definitely as an advanced one.

par Holger P

19 nov. 2017

This course is amazing. Taught by experts in the field with a proven track record of outstanding performance in Kaggle competitions. They teach how to fine tune ML models to achieve better performance. My choice for best course on Coursera!

par Sergio A G P

14 oct. 2020

It was a great, demanding, and very detailed course about machine learning and implementations in the context of competitions. Thus, the focus is very competitive and programmatic, but without forgetting the understanding of the problem.

par Steven A

30 mai 2021

Excellent course, challenging and interesting. A good mix of theory and practice, taught by a dynamic and passionated team capable of some tongue in cheek humor.

Note: don't expect interactive support from the forums, you're on your own.

par Amit K S

20 janv. 2019

This is so good. Three reasons (1) Helps me revisit the concepts that I learnt in the machine learning course. (2) Helps me to deal with my FOMO (3) I would feel most confident to go for my Data Science or Data Engineering interviews.

par ashesh g m

9 juin 2019

It was one of the best courses which I've done on coursera. Here, it was all practical and essential knowledge which was taught. Mentors were amazing and inspiring. A must do for any data scientist or aspiring data scientist.


2 nov. 2020

Excellent course, at the beginning was very frustrating, because of my poor knowledge, but then, with some dedication and new learnings, was great to know and practice new technique to solve real and competition ML problems.