FastChai and Kaggle: Group based Projects

Part 1 (2020) Study Groups
106

@Romandovega mandovega would you like to work with someone else? or are you going solo. I am also working on the same problem.

107

I am working on fraud detection. I spent a few weeks trying RF and missed some really crucial details as pointed out in the TOP solution. It is explained quite well and the goal is to get top 10% using this. Anyone will to join is welcome.

108

Great, will message you on discord.

109

Hi, I am interested. How do we meet?

1 Like
110

message me on discord please.

111

Hi,

I spent about roughly 2 months trying and understanding different solutions for a competition as part of fastchai initiative to get to top 10% solution somehow.

This is part of a series of posts divided as follows:

  1. Top 10% solution Detailed version on my blog
  2. Top 10% solution short version for kaggle
  3. What works in feature engineering
  4. How to know if you are actually overfitting
  5. What can Adverserial Validation do for you?

Cross-posted all from my DS blog. Fraud Detection Competition page is here.

Please have a look and let me know if you found any issues or if you found anything useful.

Thanks.

3 Likes
112

@neomatrix369 do you have a spot in the group?

Any other NLP groups?

113

Any luck with the implementation?