Note
You can add multiple datasets for Training. Switching between training datasets can be achieved using template parameters.
In order to train a model, a dataset needs to be provided and made available to SAP AI Core. You can achieve this by storing the datasets in a hyperscaler object store and providing the required access rights to SAP AI Core.
Hint
A hyperscaler object store represents the most cost effective, robust, and durable way of managing datasets and models since it is characterized by the ability to massively expand under a controlled and efficiently managed software-defined framework. Moreover, as SAP AI Core is designed to work with containers, an object store makes data access convenient for Argo in a transparent, automatic fashion by copying data in/out of pipeline containers and assuring secure credential storage.Step 1
In the model train block connects an object store to SAP AI Core and loads the training dataset.
Step 1 also segregates the AI artifacts, which is achieved by balancing the creation of SAP AI Core instances and resource groups. Clearly, different tenants can isolate different AI artifacts, but it is also possible to define namespaces to group and segregate related resources within the scope of one SAP AI Core tenant.
AI Core: Register the Input Dataset
One example of an asset that can be segregated within a resource group is the input dataset used to train an ML model. The registration of a dataset in SAP AI Core assumes the creation of a resource group. That resource group is only visible to those people allowed to access that specific resource group and is consumed by all those components belonging to the same resource group.
Note
Although the segregation of assets using Res.grps is valid, the important point here is to provide different datasets that are injected into the same training pipeline in different runs, however, resulting in different models.
This can be done periodically, or by shuffling one large dataset into different variants to get the best generalizing model.
AI Core: Create a Configuration
As mentioned before, you can use SAP AI Core to design their own training pipeline and specify it into a training template, which is used to define an executable object in SAP AI Core. For the sake of convenience, it falls under the umbrella of a specific scenario that acts as an additional namespace and groups all the executables together that you need to solve your specific business challenge. The executables, together with the input dataset, are the key components that you need to bind together to create a training configuration and then to train an ML model in SAP AI Core.
Note
For the sake of convenience, it falls under the umbrella of a specific "scenario" that acts as an additional namespace.
Unfortunately, the scenario is not a namespace. It is used for grouping items of a given ML "project."
A configuration is a set of parameters which can be changed for every run.
AI Core: Trigger the Workflow Execution
An executable includes information about the training input and output, the pipeline containers to be used, and the required infrastructure resources. SAP AI Core provides several pre-configured infrastructure bundles called resource plansthat differs from one another depending on the number of CPUs, GPUs, and the amount of memory. Therefore, SAP AI Core can efficiently deal with demanding workloads.
Another important point when working with ML, is evaluating the model’s accuracy. ML models are used for practical business decisions, so more accurate models can result in better decisions. The cost of errors can be huge, but the impact of these mistakes can be mitigated by optimizing the model’s accuracy. This being said, a customer using SAP AI Core might ask: how am I going to evaluate the performance of my model? The answer is that SAP AI Core provides several APIs to register your favorite metrics that can be retrieved and inspected once the training is complete.
AI Launchpad: Check the Status of Execution and Logs
When the model has completed training and has satisfactory metrics, the final step of the model training block consists of harvesting the results. When the model is produced and stored by SAP AI Core in the same connected hyperscaler object store, it is automatically registered in SAP AI Core and is ready for deployment.