Dear SAP Community,
In this edition of our quarterly What’s New blog, we are happy to introduce the latest features that were released for SAP AI Core & SAP AI Launchpad in Q2 2022.
SAP AI Core is SAP’s runtime for heavy-load AI. It allows you to train and deploy your AI models cost-efficiently at scale. Make your choice from a broad range of storage, CPU and GPU service plans and benefit from auto scaling and accelerated performance with GPU support.
SAP AI Launchpad is the central application to access and manage your AI use cases. Connect to multiple runtimes, such as SAP AI Core, and streamline the AI lifecycle management of your use cases. Leverage SAP AI Launchpad to monitor your models and continuously improve their performance.
Use SAP HANA Cloud, data lake files as an object store for your artifacts that is natively embedded in the SAP software environment! Data lake files are part of SAP HANA Cloud, data lake and are based on the WebHDFS protocol.
The screenshot above shows you what details you need to provide to connect to your instance of data lake files. With that, data lake files provide an alternative storage for your artifacts directly within the SAP software environment and grant you more flexibility in where to store your artifacts. As of today, you can leverage this new storage option to retrieve and store artifacts during model deployments. Support for executions will be added in an upcoming release.
Help: Register Your Object Store Secret
Easily manage and track your metrics using the SAP AI Core SDK! The SDK now includes a module that provides convenient, programmatic access to the metrics tracking of AI API. With that, you can log metrics during training runs and assign tags and other information. Afterwards, you can query, modify, and delete your metrics with the SDK. In sum, you can now perform full metrics tracking with the convenience of our SDK and invoke it directly in your Argo Workflows.
Help: Metrics Tracking with SAP AI Core SDK
Great, new features are available in SAP AI Launchpad to analyze metrics of your executions and models and gain deeper insights.
First, you can now compare metrics of your training executions or models in a tabular way. This allows you to optimize your business results by easily identifying the configurations producing optimum results or determining whether you need to make further adjustments.
Second, you can visualize the same metrics of executions and models in charts. Choose from different types of charts, view them fullscreen, or create multiple charts and view them in parallel. Using charts, you will be able to identify trends and differences across executions and models more easily than in tables.
Both features support comparison of metrics across up to five executions or models. Further, both functionalities are available in the Function Explorer app and the ML Operations app.
Help: Compare Model Metrics
SAP AI Launchpad now enables you to perform administrative activities for your SAP AI Core runtime. More specifically, you can manage the Git repositories that are connected to your SAP AI Core instance, the sync applications, as well as the Docker registry secrets. Simply connect your SAP AI Core instance in SAP AI Launchpad and the administrative area will pop up. With that, you can now even perform the setup procedure of SAP AI Core in SAP AI Launchpad and leverage it as your one-stop shop for AI lifecycle management while benefiting from the convenience of an UI.
Help: Administration
Easily create new connections in SAP AI Launchpad with the new service key upload! While creating a new connection in SAP AI Launchpad, simply upload a service key and the connection details are automatically filled. This allows you to create new connections easier and faster, especially if you are unsure how to fill each field. For the moment, only service keys of SAP AI Core are supported.
Help: Add Connection to SAP AI Core
The Meta API is part of the AI API and identifies what capabilities are supported by a given runtime. With that, SAP AI Launchpad now automatically adapts the user interface based on the supported capabilities. More specifically, capabilities that are not supported are hidden from the user. Therefore, users of your runtime benefit from a streamlined experience and are not confronted with capabilities that they are unable to use.
Help: Custom Runtime Capabilities using the Meta API
Finally, SAP AI Launchpad added new languages and is now available in eight languages overall, including English, German, French, Japanese, Portuguese, Russian, Simplified Chinese, and Spanish. The displayed language depends on your browser’s locale.
This blog post only provides a high-level overview of the latest features and enhancements of SAP AI Core & SAP AI Launchpad. You can find more details on these features in the SAP Help Portal and the respective pages for SAP AI Core & SAP AI Launchpad.
If you have any questions, you can comment below or post a question with the SAP AI Core or SAP AI Launchpad tag and check the blog posts about SAP AI Core and SAP AI Launchpad.
Also check out our SAP Road Map Explorer to see more upcoming features.
For more information on SAP AI Core & SAP AI Launchpad:
In this edition of our quarterly What’s New blog, we are happy to introduce the latest features that were released for SAP AI Core & SAP AI Launchpad in Q2 2022.
SAP AI Core is SAP’s runtime for heavy-load AI. It allows you to train and deploy your AI models cost-efficiently at scale. Make your choice from a broad range of storage, CPU and GPU service plans and benefit from auto scaling and accelerated performance with GPU support.
SAP AI Launchpad is the central application to access and manage your AI use cases. Connect to multiple runtimes, such as SAP AI Core, and streamline the AI lifecycle management of your use cases. Leverage SAP AI Launchpad to monitor your models and continuously improve their performance.
Contents
- SAP AI Core
- SAP AI Launchpad
- Comparison and Visualization of Metrics
- Availability of SAP AI Core Administration App
- Service Key Import
- Meta API
- Availability of New Translations
SAP AI Core
SAP HANA Cloud, data lake files as Object Store
{
"name": "default",
"type": "webhdfs",
"pathPrefix": "<path prefix to be appended with bucketname>",
"data": {
"HDFS_NAMENODE": "https://<file-container-name>.files.hdl.canary-eu10.hanacloud.ondemand.com",
"TLS_CERT": "-----BEGIN CERTIFICATE-----\nMIICmjCCAYIxxxxxxxxxxxxR4wtC32bGO66D+Jc8RhaIA==\n-----END CERTIFICATE-----\n",
"TLS_KEY": "-----BEGIN PRIVATE KEY-----\nMIIEvQIBADANBgkqxxxxxxxxxxxxnor+rtZHhhzEfX5dYLCS5Pww=\n-----END PRIVATE KEY-----\n",
"HEADERS": "{\"x-sap-filecontainer\": \"<file-container-name>\"}"
}
} Use SAP HANA Cloud, data lake files as an object store for your artifacts that is natively embedded in the SAP software environment! Data lake files are part of SAP HANA Cloud, data lake and are based on the WebHDFS protocol.
The screenshot above shows you what details you need to provide to connect to your instance of data lake files. With that, data lake files provide an alternative storage for your artifacts directly within the SAP software environment and grant you more flexibility in where to store your artifacts. As of today, you can leverage this new storage option to retrieve and store artifacts during model deployments. Support for executions will be added in an upcoming release.
Help: Register Your Object Store Secret
Metrics Tracking in SAP AI Core SDK
from ai_core_sdk.models import Metric
from datetime import datetime
ai_core_client.metrics.modify(
resource_group = "tutorial",
execution_id = "e6663ea0593fe2b2", # Change this
metrics = [
# list of Metric object
Metric(
name="Accuracy",
value=float(86.99),
timestamp= datetime.now().utcnow(), # ensure in UTC, easier for local time conversion
step = 0,
labels = []
),
]
) Easily manage and track your metrics using the SAP AI Core SDK! The SDK now includes a module that provides convenient, programmatic access to the metrics tracking of AI API. With that, you can log metrics during training runs and assign tags and other information. Afterwards, you can query, modify, and delete your metrics with the SDK. In sum, you can now perform full metrics tracking with the convenience of our SDK and invoke it directly in your Argo Workflows.
Help: Metrics Tracking with SAP AI Core SDK
SAP AI Launchpad
Comparison and Visualization of Metrics
Great, new features are available in SAP AI Launchpad to analyze metrics of your executions and models and gain deeper insights.
First, you can now compare metrics of your training executions or models in a tabular way. This allows you to optimize your business results by easily identifying the configurations producing optimum results or determining whether you need to make further adjustments.
Second, you can visualize the same metrics of executions and models in charts. Choose from different types of charts, view them fullscreen, or create multiple charts and view them in parallel. Using charts, you will be able to identify trends and differences across executions and models more easily than in tables.
Both features support comparison of metrics across up to five executions or models. Further, both functionalities are available in the Function Explorer app and the ML Operations app.
Help: Compare Model Metrics
Availability of SAP AI Core Administration App
SAP AI Launchpad now enables you to perform administrative activities for your SAP AI Core runtime. More specifically, you can manage the Git repositories that are connected to your SAP AI Core instance, the sync applications, as well as the Docker registry secrets. Simply connect your SAP AI Core instance in SAP AI Launchpad and the administrative area will pop up. With that, you can now even perform the setup procedure of SAP AI Core in SAP AI Launchpad and leverage it as your one-stop shop for AI lifecycle management while benefiting from the convenience of an UI.
Help: Administration
Service Key Import
Easily create new connections in SAP AI Launchpad with the new service key upload! While creating a new connection in SAP AI Launchpad, simply upload a service key and the connection details are automatically filled. This allows you to create new connections easier and faster, especially if you are unsure how to fill each field. For the moment, only service keys of SAP AI Core are supported.
Help: Add Connection to SAP AI Core
Meta API
The Meta API is part of the AI API and identifies what capabilities are supported by a given runtime. With that, SAP AI Launchpad now automatically adapts the user interface based on the supported capabilities. More specifically, capabilities that are not supported are hidden from the user. Therefore, users of your runtime benefit from a streamlined experience and are not confronted with capabilities that they are unable to use.
Help: Custom Runtime Capabilities using the Meta API
Availability of New Translations
Finally, SAP AI Launchpad added new languages and is now available in eight languages overall, including English, German, French, Japanese, Portuguese, Russian, Simplified Chinese, and Spanish. The displayed language depends on your browser’s locale.
This blog post only provides a high-level overview of the latest features and enhancements of SAP AI Core & SAP AI Launchpad. You can find more details on these features in the SAP Help Portal and the respective pages for SAP AI Core & SAP AI Launchpad.
If you have any questions, you can comment below or post a question with the SAP AI Core or SAP AI Launchpad tag and check the blog posts about SAP AI Core and SAP AI Launchpad.
Also check out our SAP Road Map Explorer to see more upcoming features.
For more information on SAP AI Core & SAP AI Launchpad:
- Follow us in the SAP Community: SAP AI Core | SAP AI Launchpad
- Start innovating with our tutorials: SAP AI Core | SAP AI Launchpad
- Find guidance in the SAP Help Portal: SAP AI Core | SAP AI Launchpad
- SAP Managed Tags:
