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

Challenge #94: Have we reached Peak Pumpkin?

jasperlch
12 - Quasar

Solution attached.

JORGE4900
8 - Asteroid

@SeanAdams I didn't know lizards could give such advise. 

 

Hey there Sean,

 

I think that I saw one of your posts on the Slack forum about lesson 3 of ARIMA models from Udacity. Is that you taking that certification or is someone else using the same name and last name?

 

If you are, probably you completed the certification by now; I am also enrolled into that certification and almost done.

 

Best regards,

SeanAdams
17 - Castor
17 - Castor

:-) Hey @JORGE4900 - you're absolutely right about the Udacity course - finished this about a month ago.   I must say, the team did a great job of this course ( @PatrickN) - and I'd recommend this to anyone who has not used the predictive tools.

 

Good luck with the final pieces Jorge, and thank you for mentioning this!

Sean

JORGE4900
8 - Asteroid

No wonder your name seemed familiar @SeanAdams.

 

I am about to go into the 7th lesson just before the capstone project. Have you already put into practice some of these techniques at work? I immediately started doing a linear regression model to calculate prices on some pieces of equipment at my work and modified the workflow to project dynamic monthly prices for the next 12 months depending on changing variables; however, this model has not been adopted by the company despite its accuracy and usefulness.

 

Best regards,

JosephSerpis
17 - Castor
17 - Castor

Challenge Completed

jamielaird
14 - Magnetar

Here's my solution.

 

Spoiler
It was nice to experiment with the Time Series tools as I've never used them before. It took me a while to figure out that Model Factory and Forecast Factory weren't in the standard set of predictive tools I'd installed with 2018.1 so I added them to my own macros folder.Screen Shot 2018-04-15 at 15.12.00.png

 

 

 

 

kat
12 - Quasar

I'm a little confused by this one... I went the macro route to score each product type. Based on that I chose my model and did another model to get the forecast. The scoring is already a little odd - all forecasts look alright, except for bakery which is just flat. Then when I run the forecast all numbers are flat other than produce. The same happened when I ran the solution - doesn't look anything like the solution in the start file. @ChristineB - I know this is an old challenge, but is there any reason you know of why me running your solution file on my computer could produce different results? The last screenshot below is from me running the solution workflow.

 

Spoiler
example of the scoring outputexample of the scoring outputforecast macroforecast macroscoring macroscoring macro

this is what I get when I run the solution file...this is what I get when I run the solution file...

 

ChristineB
Alteryx Alumni (Retired)

@kat thanks for bringing this to my attention!  I'll investigate...

ChristineB
Alteryx Alumni (Retired)

I've added a new start file and solution to this challenge to address the changes to Time Series calculations with the ARIMA tool since version 2018.2 (with an accompanying upgrade to the version of R used in the predictive tools).  I've attached them to this post and provided additional instructions on the challenge description. 

danilang
19 - Altair
19 - Altair

Good morning

 

@ChristineB.  It looks like you're from the future! Your solution and macros were developed with 2018.3.  I put on my CSE hat, figured out how fix the version info on all 3 and also updated the paths to the macros in the .yxmd.  The revised package is attached.

 

It looks like @kat's concerns are still valid though.  When using ARIMA, the  forecasts are still flat for most of the products.  Any idea why this might be?

 

Kat.  How are you progressing with the challenges?  This is 125 for me.

 

Dan