Datarobot delivers testing, optimization and AI observability to enable customers to create production-grade AI applications, observe and intervene in real-time and govern and optimize the infrastructure.

Create Production-Grade AI Applications

Build safe and useful generative and predictive applications with advanced RAG testing and evaluation techniques.

• Enterprise-grade open source LLM support: Leverage any open source foundation models — LLaMa, Hugging Face, Falcon or Mistral, and new models like Nematron-3-8B.
• LLM Metrics Evaluation and Assessment: Quickly evaluate the quality of your RAG pipeline with metrics like correctness, faithfulness and effectiveness along with user feedback integration and guard model testing to ensure optimal performance and safety.
• LLM Playground Advanced Testing: Advanced production-tested RAG workflow and customization tools let you test various embedding strategies, chunking strategies, and LLMs.
• Notebook Codespaces: Seamlessly collaborate on AI projects accessible from anywhere, version control with Git, work on multiple notebooks simultaneously, all in a user-friendly interface for efficient code development and deployment.
• Model Training on GPUs in Workbench: Accelerate model training and improve productivity with NVIDIA Rapids GPU accelerated libraries in DataRobot notebooks.
• Q&A Chat App: Accelerate experimentation with ready to use, interactive, GenAI application that can be shared with stakeholders to test GenAI experiments created in the DataRobot playground
• App Workshop: Eliminate complex deployment hurdles with a release-ready AI app workshop. Centralize how users register, catalog, deploy, and manage the full life cycle without tool hopping. 
• Prompt Tracing: Pinpoint the source of a model’s performance problem and map it back to the place in your vector database causing the issue, then leverage user feedback to train predictive models, enhancing model performance and user experience.

Observe and Intervene in Real-Time

Quickly detect and prevent unexpected and unwanted behaviors.

• Unified Registry for GenAI Apps and ML Models: Standardize AI visibility, deployment, integration, and monitoring for individuals or groups of machine learning models and AI applications.
• GenAI Guard Library: Control model performance with a full suite of out-of-the-box metrics, custom guard models, and methods, including resource consumption, PII detection, toxicity, faithfulness, and more.
• Real-time LLM Intervention and Moderation: Create a strong, multilayered defense strategy and minimize risk with dynamic real-time oversight and intervention methods at prompt and response to prevent issues like hallucinations, prompt injection, and PII leakage.
• Multi-Language GenAI Text Drift: Assess topic trends from user interactions and leverage data drift word cloud insights to augment vector databases, adjust RAG models, or fine-tune models for text generation projects.
• Custom Governance Tests via Jobs: Validate model performance and behaviors, and create custom explainability insights to export charts and coefficients for compliance documentation.

Govern and Optimize Infrastructure Investments

Get more value from your existing infrastructure with a purpose-built AI platform.

• NVIDIA Inference Triton Server Integration: Seamlessly deploy high-performance models with NVIDIA’s Triton Inference Server integration, with extra acceleration on all your GPU-based models, optimizing inference speed and resource efficiency.
• Optimized AI Inference with NVIDIA Inference Microserves (NIMs): Enhance model training and remove the need for individual GPU-powered systems with NVIDIA Inference Microservices in DataRobot.
• Cross-Cloud & Hybrid AI Observability: Effortlessly manage your AI portfolio across cloud and hybrid environments with comprehensive observability, cross-environment visibility, and unified governance.
• Global Models: Ensure consistent security and performance monitoring across your AI assets with Open Source Deep Learning and NLP models and share the best performing models with contributors.
• Registry Jobs and Notification Policies: Validate model performance and behavior and reduce time-to-detection and time-to-resolution with real-time notifications and highly customizable alerting. 
• Custom Apps Sharing: Safely share custom GenAI apps with stakeholders inside or outside your organization while adhering to governance and security policies through granular RBAC and governance policies.

Go the DataRobot Documentation Release Center for more information.

The post DataRobot Spring 2024 appeared first on DataRobot AI Platform.

DataRobot’s newest release gives you the confidence to achieve real-world value with generative AI, enabling you to rapidly build with optionality, govern with full transparency, and operate with correctness and control.

Operate with Correctness and Control

Continuously improve models with comprehensive insights and guardrails while maintaining data security at scale.

Unified Console: Observe all models in one unified location before production – generative and predictive, regardless of their origin.

GenAI Guard Models and Human Feedback Loop: Combine generative AI with predictive models to evaluate attributes like toxicity, sentiment, personal information leaks, prompt ingestion, and correctness for models in production. Continuously improve your models through user feedback on the quality of their generated responses.

Out of the Box and Custom Metrics: Evaluate generative and predictive models using pre-defined metrics such as text drift and service health, and configure use case-specific metrics to measure performance against desired business goals. Receive real-time notifications if metrics exceed designated thresholds to quickly intervene.

Govern with Full Transparency

Govern all AI from one unified location – generative and predictive models, built on or off-platform.

Unified Model Registry: Test and evaluate both generative and predictive models in one unified location before production, regardless of their origin.

GenAI Cost Insights Metrics: Monitor GenAI costs in real-time at the deployment level with customizable metrics for informed cost-performance decisions; receive alerts when costs surpass specified thresholds.

Enterprise-Grade LLM Deployments: Register each deployment as one flow and monitor the complete GenAI ‘recipe,’ set workflow approvals and permissions for production changes, and enable effortless version control.

Custom Job Insights: Create groups of tests with custom environments and dependencies for model approval, scenario checks and personalized explainability insights.

Batch-Based Monitoring: Track and manage model monitoring statistics, such as data drift and accuracy, by specific batch jobs.

Global/Public Models in Registry: Access public models added by DataRobot or your organization’s admin, such as a Toxicity Classifier from Hugging Face, without requiring each user to register or explicitly share them.

Notification Policies: Configure and get model performance and health notifications instantaneously.

Rapidly Build with Optionality

Rapidly deploy new GenAI use cases using an intuitive interface to experiment with the underlying components of your choice.

Multi-Provider LLM Playground: First-of-its-kind visual comparison interface with out-of-the-box access to external LLM services, including Google PaLM, Azure OpenAI, and AWS BedRock, or the ability to bring your own – to easily compare and experiment with different GenAI ‘recipes’ with any combination of foundation models, vector databases, and prompting strategies.

Model Comparison: Easily and visually compare insights and lineage across model experiments within a single use case.

Network Access for Custom Tasks: Customize your model blueprint pipelines to integrate external models or services through APIs, including your own LLMs and external embedding.

Vector Database Builder: Extend LLMs with your company data to protect data privacy and ensure actionable responses for your organization; easily track the lineage of any vector database created within our Notebook or the UI.

Enterprise Messaging App Integrations: Seamlessly integrate with the messaging apps your organization uses, like Slack and Microsoft Teams, to connect with end users and ensure widespread adoption.

Notebook Scheduling: Optimize your workflow by scheduling or triggering your notebook, all while maintaining detailed execution tracking automatically.

Notebook Custom Environment Integration: Define and reuse custom dependencies and environments within notebook sessions and custom models.

GPUs (Public Preview): Leverage automated GPU utilization when working with unstructured data, whether you’re coding or using the GUI, all while maintaining safeguards.

GenAI Accelerators: Use our pre-built AI Accelerators templates to quickly build GenAI specific industry and department use cases, or to enable LLMOPs and observability of GenAI solutions built with Google PaLM, Azure OpenAI, AWS BedRock, and more.  

Data Integration Enhancements: Seamlessly connect and wrangle data in Databricks or materialize data in BigQuery.

Advanced Modeling Settings in Workbench: Explore advanced modeling settings within the new Workbench experience.

Time Series in Workbench: Utilize our time series modeling capabilities within the new Workbench experience.

Go the DataRobot Documentation Release Center for more information.

The post DataRobot Fall 2023 appeared first on DataRobot AI Platform.

DataRobot gives you the only open solution that delivers on your generative and predictive AI needs from experimentation through production and consumption across any cloud.

Build Exceptional, End-to-end Generative and Predictive AI Solutions

Maintain choice, control, and flexibility for any generative and predictive AI use case.

DataRobot hosted Notebooks: Create vector databases, interact with your preferred LLM, and experiment with prompt strategies, all in a single notebook, from development to production.

Suite of code-first experiences and templates: Get started on AI projects with generative AI and predictive AI Accelerators, Azure OpenAI-powered Code Assist (preview), and custom Streamlit app integrations.

Deep learning for text-based AI: Enhance predictive AI use cases with generative AI. Utilize advanced DL and NLP for text, including end-to-end PDFs ingest, embedded foundational models for preprocessing, and seamless integration with Hugging Face Models.

Bring Generative AI and Predictive Insights to Your Business

Create generative AI and predictive applications that everyone in the business can easily understand.

Hosted Streamlit application: Easily prototype and customize generative AI applications with a hosted Streamlit sandbox.

Interactive Insights: Translate predictions into interactive, visually intuitive insights for your stakeholders with enhanced insights AI apps.

Deploy and Maintain Safe, High-Quality, Generative AI Applications in Production

Manage and monitor generative AI assets in production for sustained performance

Model Registry & Versioning: A centralized registry to organize, version, and deploy LLMs. Confidently upgrade your LLMs and effortlessly revert deployment to a previous version.

Monitoring Generative AI: Monitor model performance and SLAs on generative AI projects regardless of where they are deployed, or who built them. 

Custom Performance Metrics for LLMs: Track what really matters and protect business reputation with custom metrics such as toxicity monitoring, cost of LLM, or whether your LLM is staying “on-topic”.

Schedule Monitoring Jobs & Business Rules: Schedule monitoring jobs for all models without any manual pipelines or data movement. 
With Timeliness Indicator, set your own rules and alerts for how frequently predictions or actuals should run. 

Enhanced Integrations: Streamline AI Production pipelines, ingest Airflow data, run predictions on Databricks Spark clusters in 2 lines of python code, or deploy models to AzureML in a single click.

Ensure Secure, Adaptable, and Efficient AI Lifecycle Across Your Infrastructure

Deliver the ultimate outcome with your existing investments

Kubernetes VPC installation: Experience the transformative power of generative AI on your preferred cloud platform with DataRobot support for Kubernetes installation.

Safe connection to cloud warehouses: Securely access your Google and Snowflake environments using OAuth, Key-Pair Authentication, and Service Accounts while maintaining full control and security.

Maximize cloud speed and compute: Accelerate data preparation tasks like deduplication, table joins, aggregations, and write-back training data in Snowflake and Google BigQuery.

HIPAA compliant: Ensure infrastructure sovereignty with our HIPAA compliant Single-Tenant SaaS on Azure and AWS.

Accelerate Generative AI Projects and Cultivate Internal Expertise

Executive and practitioner programs to accelerate your generative AI journey

Strategic Advisory

• Generative AI for Executives: A program primer on generative AI – the why, what and how – targeted at senior leaders
• Generative AI Roadmapping: Workshop to ideate and prioritize high value generative use cases & develop a roadmap

Technical Enablement

• Hands-on Labs: Generative AI lab series and workshops to follow along with our GenAI experts
• Use Case Execution: Dedicated time with our generative AI experts to take a use case from data to deployment and extract value

DataRobot AI Platform Summer 2023 Release Full Feature List

For the full details of features included in the DataRobot AI Platform Summer 2023 Release, visit the DataRobot Documentation Release Center.

The post DataRobot Summer 2023 appeared first on DataRobot AI Platform.

Collaborative Experimentation Experience

By empowering teams to closely align with data, models, and subject matter experts, DataRobot offers Machine Learning Experimentation for model data preparation and model building. A newly designed experience interface, rethought user workflows, and an intuitive code-first and no-code/low-code experience for data science practitioners delivers faster iteration and experimentation. 

Workbench

It is a new experience for collecting and managing use cases to comprehensively bring together business problem assets in one location. Workbench simplifies collaboration by enabling resource sharing in a single click, solving the issue of resources scattered across internal locations, hard drives, and GitHub repositories.

Data Preparation

Data Preparation designed specifically for ML data preparation, streamlining one of the most tedious and essential steps in AI/ML projects. Easily and quickly analyze, and transform structured data directly from Snowflake without compromising security, compliance, or financial controls.

Notebooks

Notebooks are fully managed, hosted, and embedded in the DataRobot AI Platform, giving data scientists flexibility to leverage code snippets, pre-installed dependencies and versioning.

Value at Production Scale

Integrating production AI models with business workflows and applications, DataRobot aligns the models to be deployed near the critical data sources and business applications where decisions are made. DataRobot offers Machine Learning Production that scales value with new machine learning and automation capabilities for model testing and documentation, governance, model integration, and monitoring, using devops tools and best practices. 

Uniquely, DataRobot provides production capabilities regardless of whether the models are built in a standalone notebook or the DataRobot GUI, or where they are deployed—inside the DataRobot platform, in a data warehouse, or inside an enterprise application.

GitHub Marketplace Action for CI/CD

This is an advanced integration with custom model repositories in GitHub that supports ML Engineering teams to automate workflows, including custom (non-DataRobot) model deployments on DataRobot, while maintaining governance standards.

Custom Inference Metrics

Uniquely embed your own analytics to calculate model metrics critical to your business, including novel drift or accuracy calculations. Add your business KPIs to the rich metrics already provided by DataRobot to fully track model performance.

Drift Management

Retain control even as global market conditions continue to dramatically change on a frequent basis. Track model drift, alert when a model should be retrained, explain why models may be drifting, and visualize data drift for text features, understanding how models change over time.

Assured Compliance and Governance 

Today’s organizations need AI to be trusted, accountable, and governed. With DataRobot, you can set up automated and custom tests of model performance and automatically document any model’s behavior for compliance. With this, you can help ensure that models used in business-critical applications and workflows are managed to the highest standards, meet government regulations, and reduce overall risk from access or changes to models in production. 

Compliance documentation for external models

This feature saves data scientists hours of work by automatically creating compliance documentation for external models with the click of a button. Customization of compliance documentation helps data scientists adhere to enterprise or industry-specific requirements. Read more 

MLflow Integration

Bring metadata from MLflow to the DataRobot platform. Expanding the existing DataRobot MLOps integrations, this MLflow work supports the need for AI teams to have flexibility and interoperability while also relying on the DataRobot AI Platform as a central location to manage a suite of models.  Read more 

Bias Mitigation tools

Identify and mitigate bias in individual models, building on the existing DataRobot Bias Management tools. As governments explore bias regulations, organizations are very interested in understanding and correcting model discrimination based on features, such as race, gender, or income. 

Broad Enterprise Ecosystem 

As enterprises make substantial investments in infrastructure, practitioner tools, data platforms and business applications, the DataRobot AI Platform is an open system supporting key integrations. Working in the context of your enterprise architecture, infrastructure, data platform and business applications, the DataRobot AI Platform integrates deeply with cloud data warehouses and data lakes so that you can analyze data, complete feature engineering, and deploy and monitor models. 

Extensive Snowflake integration

The integration, announced today with our longstanding DataRobot partner Snowflake, provides new functionality that enables you to work where you feel comfortable—in Snowflake or in the DataRobot AI Platform. You can now connect safely and seamlessly to Snowflake, the data cloud, wrangle and prepare data directly in Snowflake, and generate insights before the modeling process begins. Or you can start from the DataRobot interface to deploy, manage, monitor, and govern models that live in Snowflake. Now AI Builders can complete the ML lifecycle—from data prep to predictions—without repeated configuration.

SAP joint solutions

Get the right tools to quickly and safely build, deploy, and monitor ML models with both SAP data and data sources outside of SAP. DataRobot and SAP recently announced deeper alignment to conveniently build and deploy machine learning models across the SAP technology stack and deploy those models into SAP business applications. The joint solution complements the SAP embedded AI strategy, providing the right tools to quickly and safely build, deploy, and monitor ML models with both SAP data and data sources outside of SAP. 

Kubernetes support

Kubernetes support, including Red Hat OpenShift and AKS, standardizes and simplifies installation. Kubernetes is now the standard run-time environment for cloud and on-premises infrastructure. DataRobot uses standard Kubernetes processes to automate the deployment of DataRobot services across multiple compute nodes, eliminating the time previously spent in doing this manually.  

Single-Tenant SaaS

Single-Tenant SaaS support lets you outsource the setup and management of the DataRobot AI Platform to DataRobot experts, using a robust, secure, single-tenant hosted solution available on Amazon AWS, Microsoft Azure, or Google Cloud. DataRobot can even be purchased with existing cloud credits, simplifying the procurement process.

DataRobot AI Platform 9.0 Release Full Feature List

For the full details of features included in the DataRobot AI Platform 9.0 Release, visit the DataRobot Documentation Release Center.

The post DataRobot AI Platform 9.0 appeared first on DataRobot AI Platform.

Experience

Allow Users to Upload Company Logo Image in No-Code Apps

Brand your App Builder projects and customize your No-Code AI applications by uploading your company logo on the General Configuration page for No-Code Apps. This customization provides an end-to-end experience that results in a polished, personalized app that is ready to share with your decision-makers and customers alike.

Modeling

Autopilot Quickrun Improvement in AutoTS

See 11% faster results* when choosing Quickrun Autopilot mode with DataRobot Auto Time Series. With faster iterative experimentation, users can more quickly finish each model build, adjust based on results, and launch the next experiment—all in less time. Get better results by running through more cycles of improvement.

*average in pre-release testing

Sliced Insights

Easily segment and analyze specific subsets of your data, compare insights, and get explainability on the same model. With our new sliced insights capability, you can get a deeper understanding of your business, identify key trends, make more informed decisions, and optimize your strategies. Insights include feature impact, feature effects, Lift Chart, and ROC curve.

MLOps

External Model Proxy for Challengers and Compliance

This new offering enables data scientists to leverage the DataRobot AI Platform’s powerful, out-of-the-box features, such as automated compliance documentation and challengers for models that are hosted outside of the DataRobot platform. This capability extends the flexibility and support DataRobot provides for external models, eliminating the need to move them into the DataRobot platform. Using the DataRobot interface, data scientists can easily connect to externally hosted models to generate compliance reports or use these models as challengers. This feature is available on-prem only.

Within DataRobot, data scientists can define parameters, such as keys, passwords, or other controls, for any model. This makes it easy to quickly update information related to the model without changing the actual model. For example, a change in a password can easily be updated as a parameter without manually changing the model. These capabilities simplify management, governance, and compliance across all your AI models, whether they are hosted within DataRobot or elsewhere.

Custom Metrics

Calculate and track custom metrics, including business KPIs to supplement the rich metrics already provided by DataRobot. Data scientists can build and use complex or simple metrics within the DataRobot AI Platform. Custom metrics can be fed easily into business applications to track specific KPIs. These metrics can be used in model documentation for analysis with a DataRobot API, and values can be computed retrospectively for models that were not originally created within DataRobot.

Deployment History

This feature enhances the DataRobot champion-challenger framework with a new capability to maintain deployment history even when the champion model is changed. For example, when you replace a champion model, you can now view the history of important metrics, such as accuracy or predictions over time. This allows you to compare these models, track performance indicators over time, and validate whether model replacement was the right decision. Improved visibility into deployment performance enables data scientists to make informed decisions regarding models in production, such as retraining or replacing champion models.

Ecosystem

Extending Snowflake Integration: Model Deployment, Scoring, and Monitoring

Introducing improvements and new capabilities for customers to maximize their Snowflake investment. Be more productive and take advantage of better data governance with a seamless user experience when deploying and monitoring DataRobot models to Snowflake.

When deploying a DataRobot model to Snowflake, this new seamless integration significantly improves the user experience and reduces time and effort, while eliminating user errors. When models are deployed to Snowflake, leverage Snowpark to score the data for speed and elasticity without data movement.

Additionally, a new capability allows you to seamlessly monitor and govern models that are deployed to Snowflake from the DataRobot GUI. This functionality helps you keep track of your business decisions based on predictions and actual data changes and leverage DataRobot MLOps capabilities to determine the model’s health and accuracy. 

Snowflake External OAuth

The Snowflake External OAuth configuration is now available for all customers. This feature allows the implementation of external identity providers supported by Snowflake External OAuth, without providing user and password credentials to DataRobot, allowing you to keep your security and controls, while integrating your existing tools.

Python API for Scoring Code

The new official Python API for Scoring code from DataRobot makes it easy and efficient to use DataRobot-generated models outside the platform. This API has been designed keeping in mind the popular data science libraries, thus making it as simple as passing a pandas DataFrame to the API and getting one back in return. Enterprises using Databricks Notebooks will now have a more seamless experience with DataRobot. The API will be published to PyPi, making it easy for data science practitioners to install and use as part of their usual workflow.

DataRobot AI Cloud – February 2023 Release Full Feature List

For the full details of features included in the DataRobot AI Cloud February 2023 Release, visit the DataRobot Documentation Release Center.

The post February 2023 appeared first on DataRobot AI Platform.

The new, user-friendly integration with custom model repositories with existing CI/CD tools like GitHub support ML Engineering teams’ automated workflows, while maintaining the DataRobot platform’s governance standards. 

This month, DataRobot also debuts Dedicated Managed AI Cloud on Microsoft Azure. This hosted version of the DataRobot AI Cloud platform is deployed for each customer in a dedicated and separate virtual private cloud that is operated, monitored, and maintained by DataRobot in-house experts. Already available on AWS and Google Cloud marketplaces, Dedicated Managed AI Cloud can now also be purchased on the Azure Marketplace.

Experience

New DataRobot Notebooks Streamline Code-First Development

New DataRobot Notebooks provides a fully managed, hosted platform with auto-scalable compute resources for creating and executing Jupyter-compatible notebooks. It is natively integrated with the DataRobot ecosystem to streamline code-first development across the ML lifecycle so that data science professionals spend more time on data science and less time on low-level configurations or infrastructure management. With built-in code intelligence like auto-completion, code-snippets for frequently used data science functions, and pre-installed ML libraries, data scientists are assured a holistic experience that encourages faster AI experimentation. 

Modeling

Quickrun Autopilot

See 21% faster results* when choosing Quickrun Autopilot mode with DataRobot AutoML. Get more value with the DataRobot improved Quickrun Autopilot. With faster iterative experimentation, users can more quickly finish each model build, adjust based on results, and launch the next experiment—all in less time. Get better results by running through more cycles of improvement.

*average in pre-release testing

5GB Large Dataset AutoTime Series Support

DataRobot Auto Time Series support for large datasets up to 5GB is now GA. This includes ingesting, running feature derivation, and modeling on datasets up to 5GB. The expanded capability makes it easier to sample the dataset without losing important Time Series-specific information. Additionally, the larger size supports more series for multiseries modeling.

Time Series Clustering Experience Improvements

There are now desirable experience improvements to Time Series clustering modeling for a more intuitive process. Improvement in the overall user experience includes changes in the GUI when setting up a Time Series clustering project.

Build No-Code AI Apps Directly from a Model Leaderboard

Generating a No-Code AI App for your business has never been easier, thanks to a unique new building process to the existing DataRobot No-Code AI Apps capability. In just a few clicks, you can create an application directly from the model leaderboard (summary information for each model built in a project). Streamlining this process helps generate insights quickly so that users can spend more time digging into the insights that matter most to the business and less time on application setup. In addition, it simplifies the creation process of No-Code AI Apps for people without MLOps experience. 

MLOps

GitHub Actions Custom Model Deployment

This user-friendly integration with custom model repositories in GitHub supports ML Engineering  teams’ automated workflows. DataRobot platform operations (e.g.,  retraining models, champion/challenger experiments, and custom deployments) integrate with a customer’s existing CI/CD (e.g., continuous integration and continuous development) tools for repeatability and reliability. As mature customers take a DevOps approach to data science, including CI/CD, this integration allows more automated custom (i.e., non-DataRobot) model deployments on DataRobot for code-first data scientists, while maintaining the DataRobot platform’s governance standards.

TTS/LSTM Batch Predictions Support

Make batch predictions with TTS/LSTM models with the same level of automation as other machine learning models. This provides an easier process to make predictions on TTS/LSTM (Traditional Time Series/Long-Short Term Memory Models) deployments. These types of models have many benefits, as they do not require frequent retraining and are easy to understand. However, without an intuitive way to make batch predictions, these models would not otherwise make it to production.

Visualize Feature Drift for Text Features as a Word Cloud

The word cloud is a quick and intuitive way to visualize data drift for text features for critical insight into how the input data has changed over time. It shows the terms that have drifted the least and the most using different colors and shades of colors, providing an easy interpretation for text feature drift in deployed models. This word cloud capability can be found on the data drift tab of a deployment, and is available for all text-based features that have been selected for data drift monitoring.

Enable Compliance Documentation for Models without Null Imputation

Generate compliance documentation for models that do not support null imputation using DataRobot Compliance Documentation. This is accessible as a new optional template alongside the preexisting standard Compliance Documentation template. The new Compliance Documentation without null imputation template will not include the “Sensitivity Analysis” section. This update makes the DataRobot Compliance Documentation offering inclusive of more models, (i.e. models that may not support null imputation). 

Infrastructure

Dedicated Managed AI Cloud on Microsoft Azure

Reducing time to value in deploying, upgrading, and managing the AI infrastructure, DataRobot offers Dedicated Managed AI Cloud on Microsoft Azure. This hosted version of the DataRobot AI Cloud platform is deployed for each customer in a dedicated and separate virtual private cloud and operated, monitored, and maintained by DataRobot in-house experts. Ideal for organizations with specific data sovereignty issues, DataRobot Dedicated Managed AI Cloud is available for Microsoft Azure customers and can be purchased using Azure credits. 

DataRobot AI Cloud – January 2023 Release Full Feature list

For the full details of features included in the DataRobot AI Cloud January 2023 Release, visit the DataRobot Documentation Release Center.

The post January 2023 appeared first on DataRobot AI Platform.

Improve Custom Models Environment Management

Easily manage and track custom model environments and versions through DataRobot. Environments are a powerful tool for extending DataRobot Custom Models functionality to include your favorite Data Science packages. Now you can configure the number of environments and the number of saved versions allowed for your users. These updates make the environment building process transparent and easier to manage. 

R-client updated

DataRobot now provides an up-to-date version of the R client. This version ensures alignment between the R client and version 2.29 of the Public API. It is available for install from DataRobot’s public GitHub repository and comes with a host of new features, such as DownloadDatasetAsCsv to retrieve datasets as a CSV file and GetFeatureDiscoveryRelationships to access feature discovery relationships for projects that have been released since the last R client version for 2.18. This updated R client is currently in public preview.

DataRobot AI Cloud – November 2022 Release Full Feature list

For the full details of features included in the DataRobot AI Cloud November 2022 Release, visit the DataRobot Documentation Release Center.

The post November 2022 appeared first on DataRobot AI Platform.

Learn more about Drift Over Time, which helps you with further insights to identify problems and patterns over time. With more information, you can better manage predictions. Deployment Prediction Processing Usage gives you useful details to show which predictions are delayed, why they are delayed, and the time frame so you can make adjustments as needed. 

Deployment Prediction Processing Usage

Now you can visually track prediction progress with new charts and easily manage workloads and delays. See which predictions are delayed, why they are delayed, and the time frame. With improved transparency, see hourly updates on predictions that are being processed. Know when performance metrics are ready for consumption and whether they include all or a subset of predictions. These prediction tracking charts are self-serve and user-friendly.

Drift Drill Down Plot  

Changes in customer behavior and the economy can change or drift the data that is feeding your production AI. You need to track this drift across all features over time to maintain the accuracy of your predictions. The new Drift Drill Down Plot enables you to compare data drift across multiple features (or groups of features) as well multiple time periods. This can be done for both training and scoring data. Users can conveniently change the comparisons and supplement the view with contextual information such as prediction value over time to investigate the cause of the drift. Drift Drill Down Plot is available for public preview.

DataRobot AI Cloud – October 2022 Release Full Feature List

For the full details of features included in the DataRobot AI Cloud October 2022 Release, visit the DataRobot Documentation Release Center. 

The post October 2022 appeared first on DataRobot AI Platform.

In this release, DataRobot is excited to announce that Time Series Clustering is now available for SaaS users. In addition, DataRobot also focused on improving model observability with large-scale monitoring with Python, data drift monitoring over time, prediction processing stats, and more.

Also this month DataRobot Dedicated Managed AI Cloud is available for public preview. With this model, AI Cloud is deployed for each customer in a dedicated and separate VPC.  By eliminating implementation time and resources, organizations can more quickly apply data engineering, machine learning, decision intelligence, and ML Ops capabilities.

Learn more about  Dedicated Managed AI Cloud and other capabilities only found in the DataRobot AI Cloud platform.

Large-scale Monitoring for Python

Quickly aggregate raw features and predictions into monitoring stats using the Python Library to keep an eye on model performance over time. DataRobot users who currently use the DataRobot-MLOps Python library to report prediction metrics can now also aggregate raw feature and target data using the MLOps-aggregation library. Improve productivity by summarizing necessary monitoring stats—obtain more information faster and with fewer steps.

Drift Over Time

Changes in customer behavior and preferences could be fast and drastic. Data we use for more models can change overnight. You need to react quickly to changes that impact your production AI. With the new Data Drift Over Time View, you have more systematic insights to identify problems and patterns over time. For example, you can drill into a feature that has drifted and see how its drift has changed over time and how its drift relates to other features in the model. A more comprehensive view will help you identify what causes the problem, navigate through the change much faster, and proactively address issues with your model.

Feature Discovery Support with Feature Cache Enabled in No-Code AI Apps

DataRobot No-Code AI Apps now support Feature Discovery projects with feature cache enabled. Take newly created Automated Feature Discovery enabled projects that are built off multiple datasets and build simulations, find insights in a single view, run optimizations, or create what-if scenarios in a seamless, end-to-end flow, entirely within the DataRobot platform.

As seen in the image below, users can easily identify whether features stem from their primary or secondary dataset. Additionally, feature cache is a preview feature that pre-calculates the feature discovery features, and makes them available at prediction time.

Time Series What-if Scenarios

The Time Series What-If App is a No-Code AI App built specifically for Time Series projects. While some users may be familiar with the ability to build Time Series Predictor projects within No Code Apps already, Times Series What-If support expands DataRobot capabilities, allowing you to simulate different scenarios across a specific time window to optimize output. By adjusting known-in-advance features set at the beginning of project creation, you can check the sales impact of running certain marketing campaigns or staffing a certain number of employees for your retail stores over the next seven days. You can edit multiple scenarios simultaneously, resulting in significant time reduction compared to running each scenario sequentially.

Time Series Clustering

The Time Series Clustering Capability—an out of the box solution unique to DataRobot—enables users to easily identify and group similar series across a multi-series dataset. Users previously manually ran a time series clustering technique outside the platform and then used the cluster assigned as a segmenting feature within a time series segmented modeling workflow. Now, this process is seamless and entirely contained within the DataRobot platform. For example, if you are predicting sales of shoes across stores in all of North America, then DataRobot AI Cloud platform will automatically group all stores in San Francisco and Cleveland into one cluster because it detected identical sales profiles for these two locations. Users need not be familiar with advanced concepts like Dynamic Time Warp or code savvy to use the new clustering capability.

DataRobot AI Cloud – September 2022 Release Full Feature list

For the full details of features included in the DataRobot AI Cloud September 2022 Release, visit the DataRobot Documentation Release Center.

The post September 2022 appeared first on DataRobot AI Platform.

This release focuses on significant enhancements for DataRobot MLOps, specifically in the model monitoring area. These new features help you compare and evaluate production models with new insights, create custom metrics that are important to your business, and scale your monitoring to save time. 
Learn more about these and other features only found in the DataRobot AI Cloud platform.

Challenger Insights for Multiclass and External Models

Now Challenger Insights – unique to DataRobot  – are available for external and multiclass models. You can perform comparisons efficiently when your production model is hosted outside the DataRobot AI Cloud platform. Compare, evaluate, and decide to keep or replace the model based on powerful built-in performance insights: per-class accuracy, Logloss, FVE Multinomial, and more. DataRobot provides a framework for benchmarking, testing, and analyzing retrained challenger models, giving you trust and confidence before performing a model replacement in production.

Large Scale Monitoring

More quickly obtain monitoring statistics for large payloads. Calculate statistics on your edge infrastructure and send aggregated statistics back to DataRobot MLOps to monitor data drift—no need to submit entire prediction requests to DataRobot AI Cloud Platform to get data about drift monitoring. This also means you can govern and protect your sensitive data by performing monitoring analysis near where the data lives. The large-scale monitoring functionality is currently available for the both the Java SDK and the MLOps Spark Utils Library.

Compute Custom Metrics Outside the DataRobot Platform

Go beyond existing monitoring metrics—now you can access prediction and training data to compute custom metrics outside the DataRobot AI Cloud platform to use in your external applications and tools. Extract predictions anytime to monitor and check for drift or anomalies  in the production environment of your choice. 

You can extract prediction and training data for the current model or any previously deployed models. And most importantly, you can pull data based on a prediction time window, allowing you to analyze specific periods and define your strategies for model observability.

Improved Data Connection UX

Your favorite data sources are at your fingertips. Explore, add, or manage your existing or new data sources from one centralized place. As a result, save time and ensure secure access to your data. 

With a recently improved interface, you can set up new data connections in just a few clicks and start your project right away. DataRobotalso simplified connection authentication – no need to specify a credential each time you use a data connection. Configure once and ensure secure access to your data sources.

DataRobot AI Cloud – August 2022 Release Full Feature list

For the full details of features included in the DataRobot AI Cloud August 2022 Release, visit the DataRobot Documentation Release Center. 

The post August 2022 appeared first on DataRobot AI Platform.