Business Context
A major bank uses a deep neural network for real-time fraud scoring. The model file is significantly large (several hundred MBs). The test ensures that the `check` action can handle large BLOBs without timing out or running out of memory during the validation phase.
Data Preparation
Creation of a CAS table containing a large binary object to stress-test the input parameter.
Copied!
| 1 | DATA casuser.onnx_large_fraud_model; |
| 2 | LENGTH model_blob varbinary; |
| 3 | /* Constructing a larger dummy binary object */ |
| 4 | model_blob = repeat('080012...', 10000); |
| 5 | OUTPUT; |
| 6 | RUN; |
Étapes de réalisation
1
Execute validation on the large model and capture the result table.
Copied!
| 1 | PROC CAS; |
| 2 | astore.check RESULT=chk_res / |
| 3 | onnx={TABLE={caslib='casuser', name='onnx_large_fraud_model'}}; |
| 4 | PRINT chk_res; |
| 5 | RUN; |
Expected Result
The action handles the large binary object successfully. The 'chk_res' output table is printed, showing the validation status (Success/Failure) for the large model.