Once you are logged in, the SmartPredict Graphical User Interface displays full-screen. The first instance you use it, the workspace' s layout is empty . Later on, as we create new projects, it will show a mosaic view of all of them.
The GUI is composed of:
1. a left pane/sidebar containing the following elements (represented with their respective pictogram) :
2. and next to it, a clear workspace, with a:
a project-creation button
3. On the top is the logo of SmartPredict .The User icon is on its extreme right, clicking through which leads to the user space including the:
The user space is where you can manage everything about your personal information.
The left pane contains the following elements:
Dashboard: The dashboard is the main menu of the platform. From there, you may create new projects as well as consult previous ones. Those latter are arranged according to the date of creation, from the most recent to the least recent. If you have been running a project lately, it is also where you see its current status (either it has run successfully or failed to do so).
Project: The project menu opens on the project panel wherein AI workflows will be represented by flowchart following a four-step process: build, deploy,monitor , test.
Datasets: The dataset menu displays the list of all previously uploaded datasets. It is also the place to upload and store new datasets.
Applications: The Applications menu contains SmartApps for Data Processing and Visualization .
Notebook: The Notebook is a console for customizing modules through Python codes. Use it to interact with SmartPredict by embedding your own code snippets and libraries.
Settings : it is a space for setting up the user account and for integration with GitHub version controls.
Build: The first project workspace is the build workspace.
Deploy: The deploy workspace directly follows the build workspace. Executed projects from the build workspace will directly transition to this next tab, once they are prepared for deployment.
Monitor: The monitor workspace is a dashboard tab to monitor the deployed project.
Test: the test tab is intended for testing models.
The right sidebar displays the menu of different toolsets such as modules, datasets, processing pipelines and logs.
Modules: modules are drag and droppable elements. Some are native and some can be created and added to custom modules . They are located under accordions that can be expanded or collapsed to reveal their contents.
Datasets: datasets may present themselves in csv table formats , in image format or soon, in text format. They originate from uploads . There are also preset modules . Supported data formats are: .csv, .npy, .xlsx, .xls, .json, .h5, .html, .pkl ,.joblib ,.txt
Processing Pipelines: processing pipelines are collections of processing steps to apply to datasets for feeding or training a model. They result from processed datasets.
Logs: the logs recapitulate the conditions of the running build . They show after the build sessions. They contain information such as the operation flow .