Show Menu
TOPICS×

Data Science Workspace tutorials

Adobe Experience Platform Data Science Workspace uses machine learning and artificial intelligence to create insights from your data. Integrated into Adobe Experience Platform, Data Science Workspace helps you make predictions using your content and data assets across Adobe solutions. Data scientists of all skill levels have sophisticated, easy-to-use tools that support rapid development, training, and tuning of machine learning recipes - all the benefits of AI technology, without the complexity.
To learn more, begin by reading the Data Science Workspace overview .

Sensei Machine Learning API

The Sensei Machine Learning API provides a mechanism for data scientists to organize and manage machine learning services, from algorithm onboarding through experimentation and to service deployment.
The following API developer guides are available:
  • Engines - Learn how to look up your Docker registry, create an Engine, create a feature pipeline Engine, retrieve the information for an Engine, update an Engine, and delete an Engine.
  • MLInstances (recipes) - Learn how to create an MLInstance, retrieve the information for an MLInstance, update an MLInstance, and delete an MLInstance.
  • Experiments - Learn how to create an Experiment, retrieve an Experiment or Experiment runs information, update an Experiment, and delete an Experiment.
  • Models - Learn how to register your own Model, retrieve the information for a Model, update a Model, delete a Model, create a new transcoding for a Model, and retrieve a transcoded Model's details.
  • MLServices - Learn how to create an MLService, retrieve the information for an MLService, update an MLService, and delete an MLService.
  • Insights - Learn how to retrieve the information for an Insight, add a new Model Insight, and retrieve a list of default metrics for algorithms.
To learn more and get the required values for performing CRUD operations with the Sensei Machine Learning API, visit the getting started guide .

How to use JupyterLab Notebooks

JupyterLab is a web-based user interface for Project Jupyter and is tightly integrated into Adobe Experience Platform. It provides an interactive development environment for data scientists to work with Jupyter notebooks, code, and data. This document provides an overview of JupyterLab and its features as well as instructions to perform common actions.
This guide will help you:
  • Access and understand the JupyterLab interface.
  • Understand code cells and the available kernels within JupyterLab.
  • Understand GPU and memory server configuration in Python/R.
To learn more, visit the JupyterLab user guide .

Data Access in JupyterLab Notebooks

Currently JupyterLab in Data Science Workspace supports notebooks for Python, R, PySpark, and Scala. Each supported kernel provides built-in functionalities that allow you to read Platform data from a dataset within a notebook. However, support for paginating data is limited to Python and R notebooks. This guide focuses on how to use JupyterLab notebooks to access your data.
This guide will help you:
  • Read, write, and query Platform data using Python, R, PySpark, or Scala notebooks.
  • Understand the read limitations of each notebook type.

Package source files for Docker recipe authoring

A Docker image allows you to package up an application with all the parts it needs. This includes libraries and other dependencies all in one package. The built Docker image is pushed to the Azure Container Registry using credentials supplied to you during the recipe creation workflow.
This tutorial will help you:
  • Download the required prerequisites for recipe creation.
  • Understand Docker based model authoring.
  • Build a Docker image for Python, R, PySpark, or Scala (Spark).
  • Obtain a Docker source file URL.

Import a recipe

This tutorial requires you to have a Docker source file URL. Visit the package source files into a recipe tutorial if you do not have a Docker source file URL.
The import recipe tutorials provide insights on how to configure and import a packaged recipe. By the end of this tutorial, you can create, train, and evaluate a Model in Adobe Experience Platform Data Science Workspace.
This tutorial will help you:
  • Create a set of configurations for a recipe.
  • Import a Docker based recipe for Python, R, PySpark, or Scala (Spark).
To learn more, follow the import a packaged recipe UI tutorial or the API tutorial .

Train and evaluate a model

In Adobe Experience Platform Data Science Workspace, a machine learning Model is created by incorporating an existing Recipe that is appropriate for the Model's intent. The Model is then trained and evaluated to optimize its operating efficiency and efficacy by fine-tuning its associated Hyperparameters. Recipes are reusable, meaning that multiple Models can be created and tailored to specific purposes with a single Recipe.
This tutorial will help you:
  • Create a new Model.
  • Create a training run for your Model.
  • Evaluate your Model training runs.
To get started, follow the training and evaluating a model API tutorial or the UI tutorial .

Optimize a Model using the Model Insights framework

The Model Insights Framework provides the data scientist with tools in Adobe Experience Platform Data Science Workspace to make quick and informed choices for optimal machine learning models based on experiments. The framework will improve the speed and effectiveness of the machine learning workflow as well as improving ease of use for data scientists. This is done by providing a default template for each machine learning algorithm type to assist with model tuning. The end result allows data scientists and citizen data scientists to make better model optimization decisions for their end customers.
This tutorial will help you:
  • Configure recipe code.
  • Define custom metrics.
  • Use pre-built evaluation metrics and visualization charts.
To get started, follow the tutorial on optimizing a model .

Score a model

Scoring in Adobe Experience Platform Data Science Workspace can be achieved by feeding input data into an existing trained Model. Scoring results are then stored and viewable in a specified output dataset as a new batch.
This tutorial will help you:
  • Create a new scoring run.
  • View your scoring results.
To get started, follow the score a model API tutorial or the UI tutorial .

Publish a model as a service

Adobe Experience Platform Data Science Workspace allows you to publish your Model as a service, enabling users within your IMS Organization to score data without the need for creating their own Models. This can be done using the Platform user interface or the Sensei Machine Learning API.
This tutorial will help you:
  • Publish a Model as a service.
  • Score data using a service via the Platform Service Gallery.
To get started, follow the publish a model as a service API tutorial or the UI tutorial .

Schedule training and scoring for a Model

Adobe Experience Platform Data Science Workspace allows you to set up scheduled scoring and training runs on a machine learning service. Automating the training and scoring process can help maintain and improve a service's efficiency through time by keeping up with patterns within your data.
This tutorial will help you:
  • Configure scheduled scoring
  • Configure scheduled training
To get started, follow the schedule a model UI tutorial .

Create a feature pipeline

Currently, feature pipelines are only available via API.
Adobe Experience Platform allows you to build and create custom feature pipelines to perform feature engineering at scale through the Sensei Machine Learning Framework Runtime.
This guide will help you:
  • Implement feature pipeline classes.
  • Create a feature pipeline Engine using the API.
To learn more, visit the tutorial for creating a feature pipeline .

Build a Real-Time Machine Learning application (alpha)

A combination of seamless computation on both the Hub and the Edge dramatically reduces the latency that is traditionally involved in powering hyper-personalized experiences that are both relevant and responsive. Hence, Real-time Machine Learning provides inferences with incredibly low latency for synchronous decision-making. Examples include rendering personalized web page content, surfacing of an offer, and discounts to reduce churn while increasing conversions on a web store.
This guide will help you:
  • Understand the Real-time Machine Learning architecture.
  • Understand the Real-time Machine Learning workflow.
  • Understand the current functionality for Real-time Machine Learning.
  • Provide the next steps for creating your own Real-time Machine Learning model.
To learn more, visit the Real-time Machine Learning overview .