Weekly Challenge

Solve the challenge, share your solution and summit the ranks of our Community!
IDEAS WANTED

We're actively looking for ideas on how to improve Weekly Challenges and would love to hear what you think!

Submit Feedback
We've recently made an accessibility improvement to the community and therefore posts without any content are no longer allowed. Please use the spoiler feature or add a short message in the message body in order to submit your weekly challenge.
SOLVED

Challenge #127: US Grand Prix Lap 2 - Employee Retention

Highlighted
8 - Asteroid

My Solution:

Highlighted
Alteryx Partner

As noted by others, no results....

 

Spoiler
Process:
- Calculate days of employment to earlier of date of termination or date of grand prix
Lower thread....
- Run through logistic regression model (Accuracy = 0.845)
Middle thread....
- Run same variables through decision tree with Employment_status as target variable (Accuracy = 0.893 so use this one
Upper thread....
- Filter for those still employed
- Calculate score with Decision tree as other input
- Filter for those with score_terminated >= 0.5

MySolution.PNG
Highlighted
8 - Asteroid

First time to use Predictive Tool and it's easier than using python. But not get the same answer. Maybe that is because updated version.

Spoiler
Annotation 2020-01-03 132654.png
Highlighted
8 - Asteroid
 
Highlighted
8 - Asteroid

My Effort

 

Spoiler
JamesCameron_0-1579770773366.png
Highlighted
8 - Asteroid
Spoiler
Challenge 127 2020-01-24.jpg
Highlighted
8 - Asteroid

I don't get the same answers in 2019.4, suspect because of the updated version of R or libraries.

AlexBibin_0-1582518453240.png

 

Highlighted
8 - Asteroid

Had to revert back to a previous tool version, but I finally got there

Spoiler
sonyakasenkramer_0-1583161634756.png

 

Highlighted
8 - Asteroid

Here is my solution. 

Highlighted
Alteryx Partner

Spent a looong time trying to figure out why I was getting no one over 0.5 until reverted to a previous version of the tools.

Spent even longer making the decision of which model to use automatical (parsing the accuracy out of the logistic was not straightforward, but parsing it out of the previous version of the tree was ridiculously complex).

Spoiler
dsmdavid_0-1584286003015.png