SAS Visual Data Mining & Machine Learning
SAS Visual Data Mining & Machine Learning is an integrated, collaborative solution designed to solve complex analytical problems with ease. It offers an automated modeling API that allows everyone in your organization to work in the same integrated environment, from data management to model development and deployment.
Key Features:
Automated Insights:
- Automated Insights Generation: SAS Visual Data Mining & Machine Learning automatically generates insights, helping you identify common variables across all models, the most important variables selected across models, and assessment results for all models. Natural language generation capabilities create simple language project summaries, making it easy to interpret reports.
- Collaborative Environment: Team members can add project notes to insights reports, enhancing communication and collaboration.
Empower Users with Language Options:
- Embed Open Source Code: If you don’t know SAS code, you can still use open source code within your analysis, call open source algorithms within a pipeline, and access those models from a common repository. This promotes collaboration as users can work in their preferred language.
- Python Integration: Utilize SAS Deep Learning with Python (DLPy), an open-source package on GitHub, to use Python within Jupyter notebooks. DLPy offers high-level APIs for deep learning functionalities, including computer vision, natural language processing, forecasting, and speech processing. It also supports the Open Neural Network Exchange (ONNX) for easy model movement between frameworks.
Optimize Solutions Efficiently:
- Massive Parallel Processing: SAS Visual Data Mining & Machine Learning delivers superior performance with massive parallel processing. It offers feature-rich building blocks for machine learning pipelines, allowing you to explore and compare multiple approaches rapidly.
- Machine Learning Algorithms: The tool provides a variety of machine learning algorithms, such as decision trees, random forests, gradient boosting, neural networks, support vector machines, and factorization machines. You can quickly find the optimal parameter settings for these algorithms by selecting your preferences. Complex local search optimization routines fine-tune models effectively.
- Unstructured and Structured Data: Integrated machine learning programs enable you to combine unstructured and structured data, extracting valuable insights from various data types.
- Reproducibility: Every stage of the analytics life cycle ensures reproducibility, delivering trustworthy answers and insights.
Boost Analytical Team Productivity:
- Collaborative Environment: Data scientists, business analysts, and analytics professionals work in a single, collaborative environment supporting the entire machine learning pipeline. Users can access, prepare data, perform exploratory analysis, build and compare machine learning models, create predictive model score code, and deploy models with a single click.
Reduce Latency between Data and Decisions:
- Business-Friendly Annotations: The solution provides business-friendly annotations within each node, describing the methods being run and offering information about the methods, results, and interpretation.
Interpret Models with Simple Language:
- Interpretability Reports: Standard interpretability reports are available in all modeling nodes, including LIME, ICE, Kernel SHAP, PD heatmaps, and more, with explanations in simple language generated through natural language generation.
Free Trial:
- A free, 14-day trial of SAS Viya, which includes SAS Visual Data Mining & Machine Learning and capabilities for the entire analytics life cycle, is available for users to explore the tool’s capabilities.
Use Cases:
SAS Visual Data Mining & Machine Learning is ideal for organizations looking to simplify and streamline complex analytics processes. It empowers analytics teams to collaborate effectively, generate automated insights, and find optimal solutions across a variety of machine learning algorithms.