Build and evaluate higher-quality machine learning (ML) models. Increase business flexibility by putting enterprise-trusted data to work quickly and support data-driven business objectives with easier deployment of ML models.
What is data science?
Building a machine learning model is an iterative process. In this ebook, we break down the process and describe how machine learning models are built.
Explore notebooks and build or test machine learning algorithms. Try AutoML and see data science results.
Build high-quality models faster and easier. Automated machine learning capabilities rapidly examine the data and recommend the optimal data features and best algorithms. Additionally, automated machine learning tunes the model and explains the model’s results.
Data scientists need to access data in different formats from different data sources, whether on-premises or in the cloud. Use drag-and-drop data integration and preparation tools to move data into a data lake or data warehouse, simplifying access for data scientists.
AI is more trusted when multiple contributors effectively collaborate, and machine learning tools provide explanation and evaluation of models. Oracle security tools and user interfaces enable multiple roles to participate in projects and share models. Model-agnostic explanation helps data scientists, business analysts, and executives have confidence in the results.
Enables data scientists to build, train, and manage machine learning models on Oracle Cloud using an open source Python ecosystem enhanced by Oracle for automated machine learning (AutoML), model evaluation, and model explanation.
Get up and running quickly with GPU-based environments, preconfigured with popular IDEs, notebooks, and machine learning frameworks. Easily deploy from Oracle Cloud Marketplace on your choice of compute shape.
A data science platform is more than just a good set of tools for building machine learning models. Oracle's data science platform includes a complete set of capabilities to support an end-to-end data science pipeline.
We're excited to announce the release of model deployment, enabling machine learning models to be served as HTTP endpoints, receive requests, and send responses back with the model predictions in real time.
Automated machine learning (AutoML) helps data scientists by automating algorithm selection, feature selection, and model tuning. This enables faster, more accurate results that take less compute time. AutoML also enables nonexperts to leverage powerful machine learning algorithms to build better quality models.
Oracle Database includes more than 30 high-performance, fully scalable algorithms covering commonly used machine learning techniques, such as anomaly detection, regression, classification, clustering, and more. Data already in Oracle Database does not need to be moved, reducing the data management workload for data scientists and allowing them to focus on building production models.
Use and import open source libraries and frameworks from Python and R to enable data exploration, transformation, visualization, and machine learning. These include but are not limited to: pandas, Dask, NumPy, dplyr for transformation, Seaborn, Plotly, Matplotlib, and ggplot2 for visualization, and TensorFlow, Keras, and PyTorch for model building.
Quickly deploy models for access by applications and business analysts. Models can be deployed with a REST API in a serverless, scalable cloud architecture as Oracle Functions or directly in the database.
Model explanation enables experts and nonexperts alike to understand the overall behavior of a model as well as individual model predictions. With model explanation and prediction details, it’s easy to understand the importance of features and what most influences predictions.
Access data in multiple formats (including CSV, Excel, and JSON), multiple sources (including object storage, Oracle Database, MongoDB, PostgreSQL, and Hadoop), and multiple locations (on premises, Oracle Cloud, and other clouds).
Data scientists can develop data science and machine learning solutions using the most popular languages, including Python, R, and SQL. Organizations achieve better and faster results when data scientists have the flexibility to use the languages best suited to particular tasks.
Try out tools for building machine learning models. No need to sign up for a cloud account.