Community Spring Cleaning week is here! Join your fellow Maveryx in digging through your old posts and marking comments on them as solved. Learn more here!

Weekly Challenges

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

Also available in | Français | Português | Español | 日本語
IDEAS WANTED

Want to get involved? We're always looking for ideas and content for Weekly Challenges.

SUBMIT YOUR IDEA
SOLVED

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

kelly_gilbert
13 - Pulsar

I'm in the People Analytics field, so this was up my alley!

 

Spoiler
I do want to point out that we would usually train the model using a subset of the data (e.g. the state of the workforce at the end of a prior year), rather than both training and scoring the model using the same set of current employees.

Capture.PNG
jssandom
8 - Asteroid

Model output seems to be different in this version but here's my solution!

pasccout
8 - Asteroid

Here is my solution... took me 15-20 minutes to figure it out...

 

the problem is that the results are not good... I did download solutions (After) from others that look similar and they also don't provide the correct join... any special config I need to do?

jamielaird
14 - Magnetar

Here's my solution. I was only able to replicate the desired output after I changed the Decision Tree back to version 1.0.

 

Spoiler
Screenshot 2018-12-30 at 00.10.06.png
RolandSchubert
16 - Nebula
16 - Nebula

My solution - I had to change the tool version for the Decision Tree tool to 1.0 as well,
using the most current version returns no records with X_Terminated >= 0.5

Natasha
9 - Comet
Spoiler
Challenge_127.png
ipeng
8 - Asteroid

Like to do prediction

JoshuaGostick
11 - Bolide

My solution :)

Spoiler
challenge_127.PNG
cplewis90
13 - Pulsar
13 - Pulsar
Spoiler
Challenge 127.PNG
KMiller
8 - Asteroid

The solution is attached. This is weird though, my solution is the same as everyone else's but no employees have >0.5 leaving probability. I have downloaded a few other people's solutions and when I run them they do not have >0.5 probabilities either. I wonder if the latest version has changed the basis of calculation for the decision tree tool.