Use case 6 : Churn Prediction Advanced Machine Learning and Custom Code in Dataiku DSS

[gerryleonugroho]

So, I've been trying to follow along with the Use Case found here:

- https://dataiku.teachable.com/p/use-case-churn-prediction

And when I got to this part, here 

- https://dataiku.teachable.com/courses/56116/lectures/831591

Where it supposed to create a Scikit-Learn (Python) model, and it returned me the following errors:

[2020/05/30-10:57:17.794] [FRT-36-FlowRunnable] [WARN] [dku.fs.local] - File does not exist: /Users/XXXX/Library/DataScienceStudio/dss_home/managed_datasets/CHURNPREDICTION/model_scikit
[2020/05/30-10:57:17.796] [FRT-36-FlowRunnable] [INFO] [dku.datasets.file] - Building Filesystem handler config: {"connection":"filesystem_managed","path":"CHURNPREDICTION/train","notReadyIfEmpty":false,"filesSelectionRules":{"mode":"ALL","excludeRules":[],"includeRules":[],"explicitFiles":[]}}
[2020/05/30-10:57:17.800] [FRT-36-FlowRunnable] [INFO] [dku.datasets.file] - Building Filesystem handler config: {"connection":"filesystem_managed","path":"CHURNPREDICTION/train","notReadyIfEmpty":false,"filesSelectionRules":{"mode":"ALL","excludeRules":[],"includeRules":[],"explicitFiles":[]}}
[2020/05/30-10:57:17.834] [FRT-36-FlowRunnable] [WARN] [dku.code.projectLibs] - External libraries file not found: /Users/XXXX/Library/DataScienceStudio/dss_home/jobs/CHURNPREDICTION/Build_model_scikit_2020-05-30T03-57-15.627/localconfig/projects/CHURNPREDICTION/lib/external-libraries.json
[2020/05/30-10:57:17.842] [FRT-36-FlowRunnable] [INFO] [dku.code.projectLi

Can anyone point out to me what went wrong with my project instance? Thanks everyone.

Cheers,

GLN

[Alex_Reutter · Accepted solution]

Hi @gerryleonugroho, note that:

1. the Teachable materials are deprecated; I would advise sticking to Academy unless you're ready to work through issues with somewhat out-of-date materials like this.
2. the messages you've shown are warnings and not errors.  I suspect, however, that your error is "

   <type 'exceptions.ImportError'>: No module named grid_search

   "
   and you need to replace
   
   from sklearn.grid_search import GridSearchCV
   
   with
   
   from sklearn.model_selection import GridSearchCV
   
   in the code to make it work.

[gerryleonugroho]

Ah okay, in that case I just need to replace the code. Thanks @Alex_Reutter 

GLN