Primary competition visual

DSN Pre-Bootcamp Hackathon: Expresso Churn Prediction Challenge by Data Science Nigeria

Helping Nigeria
Knowledge
Completed (over 5 years ago)
Classification
Prediction
671 joined
358 active
Starti
Aug 08, 20
Closei
Aug 22, 20
Reveali
Aug 22, 20
A simple Starter Notebook using Logistic Regression
Notebooks · 13 Aug 2020, 02:17 · edited 3 months later · 12

I hosted it on GitHub

https://github.com/JUGG097/DSN-Expresso-Churn-Hackathon

You can check it out

It got a Public Leaderboard Score of 0.302

Discussion 12 answers

Thank you going through the code to learn my mistakes😓

13 Aug 2020, 03:44
Upvotes 0

Guy your code was badass leanrt alot🥳🥳. Especially the pipeline model,I have questions though on the ending part.Why did you fit the pipeline to the entire dataset instead of to the train and test data

Once again you be boss😎

13 Aug 2020, 03:59
Upvotes 0

You can as well train on only "X_train", but training on the whole X before final prediction tends to work better

Thanks for this. It helped alot

13 Aug 2020, 13:50
Upvotes 0

Please i have a question. i keep getting a value error "array length does not match index length" when i try to export to csv

13 Aug 2020, 15:40
Upvotes 0

A sample of the code will help to get an idea of the issue

pd.DataFrame({"user_id": test["user_id"], "CHURN": prediction}).to_csv("starter-submission.csv", index = False)

Thats the code i am trying to run

Cross check your "prediction" variable

So I just did. I checked for the shape of my predictions and test[‘user_id’] and the length is different and I don’t really know why

Check your prediction variable hope it not just predictions on the validation set

Just did. Cant seem to find the problem.

Hmm...lemme see how you made the