avatarEbrahim Haque Bhatti

Summary

The provided web content is a curated list of 50 insightful Kaggle discussions that offer valuable tips, tricks, and resources for data science and machine learning enthusiasts, ranging from feature engineering techniques to career advice, shared by top Kaggle Grandmasters.

Abstract

The webpage presents a compilation of expert discussions from Kaggle Grandmasters, focusing on various aspects of data science and machine learning. It includes detailed threads on advanced feature engineering methods, helpful resources for learning and project development, and strategic advice for improving performance in competitions and career development. The list covers a wide range of topics, from avoiding overfitting and ensemble techniques to tutorials on specific machine learning concepts like embeddings, CNN input size, and Bayesian optimization. These discussions are aimed at both beginners and experienced practitioners looking to enhance their skills and knowledge in the field.

Opinions

  • Feature engineering is crucial for improving model performance, with top discussions highlighting common mistakes to avoid and specific techniques for different types of data and problems.
  • The resource section provides a wealth of knowledge for those at any stage of their data science journey, including cheat sheets, interview questions, and project ideas.
  • Tips from Grandmasters include advice on avoiding overfitting, the importance of not averaging in ensemble models, and the benefits of temperature shaping.
  • Tutorials offered by the community cover complex topics made accessible, such as the use of roBERTa in TensorFlow, LSTM feature importance visualization, and the choice of k in k-fold cross-validation.
  • The list emphasizes the value of Kaggle as a platform for learning, collaboration, and professional growth in the data science community.

50 of the most profound Kaggle Discussions (tips, tricks, resources) by the top Kaggle Grandmasters

Photo by Andrea Piacquadio from Pexels

Feature Engineering

  1. Top six mistakes in feature engineering
  2. Feature Engineering Techniques
  3. Tricks for Image Classification / Regression with Neural Networks
  4. How to score 97%, 98%, 99%, and 100% in MNIST
  5. How to score over 82% Titanic
  6. 9 Computer Vision Tricks to Improve Performance
  7. 7 More Computer Vision Tricks to Improve Score
  8. Not all the features are useful but some are “golden”

9. All you wanted to know about feature engineering and you were too afraid to ask

Helpful Resources

10 How to Learn Data Science on Kaggle

11. How to Become a Data Scientist at Your Own

12. 40 Interview Questions asked at Startups in Machine Learning / Data Science

13. Get Started with Data Science

14. Data Science Cheat Sheets (all of them in one place !)

15. Statistics Cheat Sheets in a single PDF !

16. Worth Seeing Posts and Notebooks

17. Best Data Science Projects for Beginners (Datasets)

18 Best of Kaggle Notebooks #2 — Time Series

19. Top 10 Starter Online Machine Learning Courses for Beginners! [3rd Update]

20. Compiled list of 600+ Q&As for Data Science interview prep

21. Plotly & Dash => Beginner’s Guide to Dashboards

Tips

22. Advice to newbies

23. Some tips to avoid overfitting

24. Is Kaggle on the resume helping?

25. Career advice from someone you probably shouldn’t listen to

26. My Kaggle Lessons Learned in 2021

27. Techniques to improve your leaderboard position

28. 9 General Kaggle Tips

29. How to Win Kaggle Competitions

30. TPS competitions — my lessons learned in 2021

31. Better than median

32 What “overfitting the public lb” means

33. Why rounding improves the score

34 Memory Trick — Reduce Memory 8x or 16x!

35. do not average in ensemble, use temperature shaping

36. [Productivity Guide] — How to work faster with notebooks

37. Neural Network Training Tips Sharing

38. What are you working on? A place to share your projects

Tutorial

39. Embeddings, Cosine Distance, and ArcFace Explained.

40. CNN Input Size Explained

41. Tutorial on Bayesian Optimization for GBMs (for starters)

42. Grandmaster Series on NLP

43. Augmentation for text

44. QDA Explained

45 Memorization GAN Explained

46. TensorFlow roBERTa Explained

47. How to Convert Characters, Tokens, and Words

48 How To Display LSTM Feature Importance

49. Choice of number of k in k-fold cross-validation

50. Keras/PyTorch pretrained segmentation models

For more Kaggle Grandmasters’ discussions:

https://readmedium.com/50-more-profound-kaggle-discussions-tips-tricks-feature-engineering-resources-by-the-top-1cd52b031831

Kaggle
Machine Learning
Data Science
Artificial Intelligence
Learning
Recommended from ReadMedium
avatarSarowar Jahan Saurav
20 Advanced Statistical Approaches Every Data Scientist Should Know 🐱‍🚀

Data science is a multidisciplinary field that combines mathematics, statistics, computer science, and domain expertise to extract…

9 min read
avatarEmmanuel Ikogho
Data Science is dying; here’s why

Why 85% of data science projects fail

4 min read
avatarMatt Chapman
The Portfolio that Got Me a Data Scientist Job

Spoiler alert: It was surprisingly easy (and free) to make

10 min read
avatarAustin Starks
I used OpenAI’s o1 model to develop a trading strategy. It is DESTROYING the market

It literally took one try. I was shocked.

8 min read
avatarZach Quinn
3 Data Science Projects That Got Me 12 Interviews. And 1 That Got Me in Trouble.

3 work samples that got my foot in the door, and 1 that almost got me tossed out.

7 min read
avatarEgor Howell
Data Side Hustles You Can Start Today

How you can monetise your data science skills for extra income

8 min read