Oracle Cloud Infrastructure (OCI) Data Science

Open source machine learning frameworks

Dozens of popular open source tools and frameworks are included to provide familiarity and versatility for data scientists. Build machine learning models with TensorFlow, PyTorch, or add other frameworks of choice.

JupyterLab interface

Built-in, cloud-hosted JupyterLab notebook environments allows teams of data scientists to build and train models with a familiar user interface.

Data visualization tools

Popular open source visualization tools such as Plotly, Matplotlib, and Bokeh help data scientists visualize and explore data.

Data access and exploration

Oracle’s Accelerated Data Science library is a Python library that contains a comprehensive set of data connections, allowing data scientists to access and use data from many different data stores to produce better models.

Automated Machine Learning (AutoML)

The Accelerated Data Science library supports Oracle’s own AutoML, as well as open source tools such as H2O 3 and auto-sklearn. Oracle’s AutoML offers automated feature selection, adaptive sampling, and automated algorithm selection. These features, along with hyperparameter tuning ultimately generate an accurate model candidate saving the data scientist significant time.

Model evaluation

Automated evaluation generates a comprehensive suite of evaluation metrics and visualizations to measure model performance against new data and compare model candidates to make it easier for the data scientist to produce a high-quality model.

Model explanation

Accelerated Data Science model explanation includes global and local explanations to help explain the overall behavior of a model, as well as specific model predictions. For model consumers, automated model-agnostic explanations improve understanding and trust, address regulatory needs, and increase the speed of machine learning adoption.

• Faster Machine Learning with Oracle Accelerated Data Science Library

Oracle Functions

Easily deploy data science models as Oracle Functions—a highly-scalable, on-demand and serverless architecture on Oracle Cloud Infrastructure that simplifies deployment for data scientists and infrastructure administrators.

Model catalogs

Team members use the model catalog to preserve and share completed machine learning models and the artifacts necessary to reproduce, test, and deploy them.

Reproducibility and auditability

Conda Environments and model catalog features allow organizations to reproduce the original model code, library, and training dataset dependencies. This allows data scientists to re-train, reproduce, and audit machine learning models.

Shared projects

Team members use projects to organize, enable version control, and reliably share all their work, including data and notebook sessions.

Security policies

Team-based security policies allow users to include team members in projects. These policies control access to models, code, and data to facilitate collaboration but also to protect work. Security controls are fully integrated with Oracle Cloud Infrastructure Identity and Access Management.

Version control

Users connect to their organization’s Git repository to preserve and retrieve machine learning work.

Flexible data access

Any data source in any cloud or on-premises can be accessed and used by data scientists for building machine learning models providing more potential data features that lead to better models.

Self-service, on-demand compute and storage

Users select the amount of compute and storage resources they need to tackle projects of any size without worrying about provisioning or maintaining infrastructure.

Powerful hardware, including graphics processing units (GPUs)

Data scientists can build and train deep learning models in much less time using GPUs in notebook sessions. Oracle Cloud Infrastructure Data Science offers support for NVIDIA P100 and V100 GPUs.