dotnet / dotnet/machinelearning-samples
Create Sample app using a large SQL Server database
- Dominant language
- PowerShell
- Stars
- 4.7k
- Forks
- 2.7k
- Avg merge
- 2d 22h
- Merged PRs (30d)
- 1
Description
Add Large dataset stored in SQL server for Database sample instead of sql lite and small dataset. -->i.e. Use the Credit card Fraud detection dataset. Export it to SQL server and use the existing Binary Classification algorithm.
Current code to evolve:
https://github.com/prathyusha12345/machinelearning-samples/tree/SQLDbSample/samples/csharp/getting-started/LargeDatasetsInSqlServer
Related URLs about the DATASET:
https://blogs.technet.microsoft.com/machinelearning/2015/04/01/now-available-on-azure-ml-criteos-1tb-click-prediction-dataset/
https://labs.criteo.com/2013/12/download-terabyte-click-logs/
Related Issue:
https://github.com/dotnet/machinelearning/issues/3868
Shared folder: `\\ct01\data\Criteo\Spark`
Additional: Test it with AZURE SQL DATABASE HYPERSCALE: 👍
https://azure.microsoft.com/en-us/updates/azure-sql-database-hyperscale-support-for-single-databases-is-now-available/
Contributor guide
Assessment
This issue has not been assessed yet.