DP-3014 Implementing a Machine Learning solution with Azure Databricks

Price
Net: €
VAT: €

Price
Price on Request

Duration
1 day

For companies and job seekers:
this course is 100% fundable!
 

Location

Course Language
English

Training Solutions
Online Live

Request a consultation Add to favorites Print as PDF

Scalable machine learning solutions are created through the interaction of data engineering, modeling, and cloud technology. Azure Databricks is establishing itself as a central tool for modern AI architectures.

Key topics

  • Building Databricks environments in Azure.
  • Data integration and transformation.
  • Development, training, and evaluation of ML models.
  • Using MLflow for transparency and traceability.
  • Automated model deployment
  • Operation, maintenance, and optimization.

Prerequisites
Solid foundation in data analysis, scripting languages, and understanding of cloud services and machine learning.

Target audience
Data scientists, data engineers, and IT professionals with a focus on AI-powered cloud solutions.

The content promotes a holistic understanding of machine learning in Azure and supports the professional development of robust, scalable AI systems.

Print as PDF
Course content
  • Start Azure Databricks
  • Identify Azure Databricks workloads
  • Understanding key concepts
  • Discover Spark
  • Create clusters
  • Use Spark in notebooks
  • Use Spark for data files
  • Visualize data
  • Understanding the principles of machine learning
  • Machine learning in Azure Databricks
  • Prepare data for ML
  • Train a model for ML
  • Evaluate ML model
  • MLflow features
  • Experiments with MLflow
  • Model registration and deployment
  • Optimize hyperparameters with Hyperopt
  • Evaluate Hyperopt experiments
  • Scale Hyperopt experiments
  • Automation of model training.
  • Using AutoML in Azure Databricks
  • Scripts for AutoML workflows
  • Understanding deep learning
  • Train models in PyTorch
  • Distributing PyTorch training with Horovod

Frequently Asked Questions

  • Azure Databricks combines data processing, machine learning, and collaboration on a single platform. This makes it possible to prepare data, develop models, and efficiently deploy production-ready AI solutions in Azure.
  • Apache Spark processes large volumes of data in parallel and serves as the foundation for many machine learning workflows in Azure Databricks. This enables fast analysis and scalable model training.
  • MLflow documents experiments, manages models, and supports their deployment. This ensures that development steps remain traceable and that machine learning models can be further developed in a controlled manner.
  • AutoML automates tasks such as model selection, training, and hyperparameter optimization. This makes it possible to create initial machine learning models more quickly and to efficiently compare different approaches.
  • Choosing the right hyperparameters has a significant impact on a model's quality. Azure Databricks supports automated optimization to systematically improve performance and prediction accuracy.
  • This training course is designed for data scientists, AI engineers, and data engineers who want to develop, train, and deploy machine learning models in Azure Databricks.
  • A foundation in data analysis, scripting languages such as Python, and knowledge of cloud technologies and machine learning are recommended. Experience with Apache Spark is helpful but not required.
  • After completing the training, participants will be able to implement machine learning workflows in Azure Databricks, train, evaluate, and optimize models, and manage them effectively using MLflow. In addition, AutoML and deep learning features will be applied in practical scenarios.
Berlin
Bremen
Dortmund
Dusseldorf
Frankfurt
Hamburg
Hanover
Cologne
Leipzig
Munich
Nuremberg
Stuttgart

Do you have any further questions? Please contact us.