Bring Your Own Model (BYOM) lets you connect models built in external platforms with
Data 360. BYOM provides direct access to Salesforce data in external ML
platforms and uses a zero-copy approach, which eliminates the need to duplicate data. After you
connect a model, you can easily consume predictions. Creating, training, and deploying models
requires coding expertise for setup in the external platform, but connecting the models to
Salesforce is simple.
Required Editions
- Use Amazon SageMaker Models
You can build, train, and deploy custom machine learning (ML) models with Amazon SageMaker and data in Salesforce. After registering a model in SageMaker, define your prediction criteria and connect the model with Data 360 to get predictions and insights for your use case. Use the insights to enrich customer profiles, create segments, and personalize experiences across various channels.
- Use Google Cloud Vertex AI Models
You can build, train, and deploy sophisticated custom machine learning (ML) models with Google Cloud Vertex AI and data in Salesforce. After registering a model in Vertex AI, define your prediction criteria and connect the model with Data 360 to get predictions and insights for your use case. Use the insights to enrich customer profiles, create segments, and customize experiences across various channels.
- Use Databricks Models
You can build, train, and deploy custom machine learning models with Databricks and data in Salesforce. After registering a model in Databricks, define your prediction criteria and connect the model with Data 360 to get predictions and insights for your use case. Use the insights to enrich customer profiles, create segments, and personalize experiences across various channels.
- Model Endpoint Formats
To connect a predictive model and get inferences, you must use the relevant endpoint request and response formats.
