Key features
This section is used to highlight and to describe SmartPredict' s key features.
1. All-in-one AI platform
SmartPredict is an all-in-one integrated AI platform especially aimed at Data scientists, Data engineers , Data analysts, Machine Learning engineers , AI practitioners, Academics and Software engineers. It has been designed to optimize the process of AI modeling in order to increase productivity, time gain and effectiveness.
For modeling AI projects, SP offers a comprehensive environment to complete a project integrally, right from the same place.
The distinct AI modeling stages : build, deploy, monitor, and test can all four be performed within SmartPredict' s generic platform.
In support to that , thanks to a rich palette of versatile modules or specialized ones, SP also covers a large number of use cases of various levels of complexity.
2. Drag and drop workspace
SmartPredict' s mode of assembling flowcharts is easy and intuitive. Simply drag and drop modules directly to the workspace to constitute an AI visual workflow.
In the same fashion, upload all kinds of functional items such as datasets, models and processing pipelines and transport them into your workspace through drag and drop.
3. SmartApps
SmartApps are add-on utilities to enhance SmartPredict' s core. You will find them practical for integration of specific modeling purposes.
For instance, for handling respectively the purposes of labeling image datasets and the tasks related to processing data of all sorts, SmartPredict possesses two special applications:
-Image labeling
"Image labeling" is a SmartApp which makes it easy to shape labeled image datasets for your image-based machine learning applications. Open the app, create your collection and begin attaching classes or text to bounding boxes. Use the collection as a drag and drop module.
-Data Processing and Visualization
"Data Processing and Visualization" is another SmartApp designed to help you get the most out of your data by exploring, processing and visualizing them.
Apart from profiling your data; this simple to use application also deals with the conception of processing pipelines . To do so, it contains an embedded powerful tool for handling missing data, for array reshaping, sorting and normalizing and even a one hot encoder:
Just like SP's other workspaces , it also comprises a drag and drop interface for producing insightful data visualizations.
4. Notebooks for project customization
SmartPredict contains a handy set of functions for personalizing your modules. They are gathered under the Notebook applications menu.
Thus, if a particular project requires you to code your own modules, do not ever feel limited: you will be able to do so through the Notebooks.
The Notebooks allow you to extend the capacities of your projects by coding and adding custom modules, user-defined functions and snippets .
5. Integrations
In addition to its inferential capacities with the REST API Web Services , SmartPredict also offers the possibility of connecting with version control repositories such as GitHub .
Last updated
