Comprehensive Machine Learning Platform
BigML provides a selection of robustly-engineered Machine Learning algorithms proven to solve real world problems by applying a single, standardized framework across your company. Avoid dependencies on many disparate libraries that increase complexity, maintenance costs, and technical debt in your projects. BigML facilitates unlimited predictive applications across industries including aerospace, automotive, energy, entertainment, financial services, food, healthcare, IoT, pharmaceutical, transportation, telecommunications, and more.
- Supervised Learning: classification and regression (trees, ensembles, linear regressions, logistic regressions, deepnets), and time series forecasting.
- Unsupervised Learning: cluster analysis, anomaly detection, topic modeling, association discovery, and Principal Component Analysis (PCA).
Interpretable & Exportable Models
All predictive models on BigML come with an interactive visualization and explainability features that make them interpretable. They can be exported and used to serve local, offline predictions on any edge computing device or be instantaneously deployed as part of distributed, real-time production applications.
- Interpretable: Visualizations such as Partial Dependence Plots effectively generate and display thousands of model predictions at a glance while Prediction Explanations and Field Importances shed light on factors driving individual predictions.
- Exportable: BigML models are fully exportable via JSON PML (and PMML) and can be used from all popular programming languages. This means you can seamlessly plug your models into your web, mobile or IoT applications or services such as Google Sheets, Amazon Echo, Zapier, and more.
Share your Machine Learning resources using granular team and project management capabilities. BigML is a transparent, collaborative platform for all members of your organization, from analysts and developers, to engineers and executives.
- Organizations: Effectively adopt Machine Learning across your entire corporate structure. With BigML organizations, the Dashboard is a shared workspace where users can access the same projects and resources with specific roles and permissions.
- Projects: All resources in an organization exist within projects, which can be public or private. Assign user permissions according to the needs of each project, allowing other members to manage, create, or simply view resources.
Programmable & Repeatable
As an "API-first" company, BigML brings every new feature first to the REST API. Bindings & libraries are available for all popular languages, including Python, Node.js, Ruby, Java, Swift, and more.
- Reproducible: BigML's granular record keeping and transparency are crucial to meet regulatory and audit compliance requirements yet are often completely overlooked in other Machine Learning tools.
- Traceable: All resources on BigML are immutable and stored with a unique ID and the creation parameters, which enables you to track any Machine Learning workflow at anytime.
Rapidly bring your predictive modeling tasks to production through effective automation. BigML turns the difficult, time-consuming work of hand-tuning models or executing complex workflows into one-click menu options or single API calls.
- OptiML: Automatic optimization for model selection and parameterization of classification and regression algorithms saves you a lot of time by creating and evaluating hundreds of models to find the best performing ones.
- WhizzML: A domain-specific language for automating complex workflows, implementing high-level Machine Learning algorithms, and easily sharing them with others.