Business Context
A telecommunications company wants to reduce customer attrition. The Data Science team needs to build a Deep Neural Network (DNN) to predict the probability of a customer leaving based on their usage patterns and demographics. This scenario validates the basic initialization of the model architecture container.
About the Set : deepLearn
Creation and training of deep neural networks.
Discover all actions of deepLearn Data Preparation
Simulation of customer demographic and usage data.
Copied!
| 1 | DATA casuser.telecom_churn; |
| 2 | INPUT customer_id age tenure usage_min bill_amt churn_flag; |
| 3 | DATALINES; |
| 4 | 101 34 12 450 65.5 0 |
| 5 | 102 28 3 120 30.0 1 |
| 6 | 103 45 60 800 95.2 0 |
| 7 | 104 50 120 150 40.0 0 |
| 8 | ; |
| 9 | RUN; |
Étapes de réalisation
1
Initialization of the DNN model table named 'churn_model'.
Copied!
| 1 | PROC CAS; |
| 2 | DEEPLEARN.buildModel / |
| 3 | modelTable={name='churn_model', replace=true} |
| 4 | type='DNN'; |
| 5 | RUN; |
2
Verification of the model table creation using tableInfo.
Copied!
| 1 | PROC CAS; |
| 2 | TABLE.tableInfo / |
| 3 | name='churn_model'; |
| 4 | RUN; |
Expected Result
The action should successfully create an in-memory table named 'churn_model' with 0 rows and 0 columns (structure only). The tableInfo action confirms its existence in the active caslib.