Quick start guide
RapidCanvas 101: A Quick Start Guide
1. Getting Started
How to Sign In
Enrollment: Upon enrolling, you will receive an email with a sign-in portal link.
Bookmarking: Save the staging login page for quick access.
2. Navigating the RapidCanvas Dashboard
Key Features
User Profile and Settings:
Manage personal settings
Access support
Select Environment:
Dedicated workspace for AI solutions and data projects
Projects Overview:
Manage and review ongoing projects
Data Connections:
Integrate various data sources
Notebook:
Write and execute code interactively
DataApps:
Build custom dashboards for data visualization
Artifacts:
Store saved outputs or resources for future reuse
3. Connecting Your Data to RapidCanvas
Steps to Connect Data
Select a Data Source Type:
Options include databases, cloud storage, APIs, or flat files.
Enter Connection Details:
Provide the necessary credentials or upload files.
Validate and Map Data:
Use built-in tools to ensure data integrity and proper field mapping.
Test and Save:
Confirm that the data loads correctly and save the connection configuration.
4. Workspaces and Projects
What is a Workspace?
A Workspace is a dedicated space designed for an organization, team, or project, ensuring isolation to maintain data security and compliance.
Key Features of a Workspace
Custom Settings
Data Security
Access Control
How to Work in a Workspace
Access: Log in and enter your dedicated workspace.
Organize: Structure your data, projects, and resources.
Build: Start developing your AI solutions within a secured environment.
What is a Project?
A project is a structured workspace for managing all components of building an AI solution or performing an analysis.
Key Components of a Project
Data Integration
Model Development
AI Agents
Visualizations
Reports
5. Building an AI Solution in RapidCanvas
Step-by-Step Process
Create a New Project:
Click the “+” button to start a new project.
Name your project and select the appropriate environment.
Connect Your Data:
Integrate your data sources as described in the “Connecting Your Data” section.
Prepare and Explore Data:
Use the AskAI tool to analyze, clean, and sort data.
Choose Your AI Model:
Select from prebuilt models or create a custom model.
Train the Model:
Run the model and refine it based on performance.
Test and Refine:
Evaluate model accuracy and make necessary adjustments.
Deploy and Automate:
Make your model accessible for use and automate workflows.
Monitor and Optimize:
Track performance metrics and adjust as needed.
6. Recipes in RapidCanvas
What is a Recipe?
A recipe is a predefined sequence of steps and data configurations that standardizes the process of building an AI solution.
What a Recipe Does:
Automates Workflows
Ensures Consistency
Simplifies Complex Tasks
How to Build a Recipe
Choose Recipe Type:
AI-Assisted Recipe: Uses a conversational AI guide.
Rapid Model Recipe: Auto-generated by RapidCanvas, but customizable.
Code Recipe: Manually connect to an API.
Edit the Recipe:
Select and customize steps such as data cleaning, model selection, training, evaluation, and visualization.
Save and Run:
Save your recipe for future use.
Run the recipe manually or schedule/automate its execution.
Monitor Results:
Track outputs and performance.
7. Models in RapidCanvas
Different Types of Models and Their Use Cases
Binary Classification Model:
Purpose: Predicts outcomes in one of two categories.
Example: Predicting if a visitor will make a purchase.
Use Cases: Targeted marketing, fraud detection, automation.
Regression Model:
Purpose: Predicts continuous outcomes.
Example: Estimating how temperature affects lemonade sales.
Use Cases: Forecasting sales, planning, understanding relationships.
Multiclass Classification Model:
Purpose: Predicts outcomes with more than two categories.
Example: Recommending items in a clothing store.
Use Cases: Personalized recommendations, organizing data, enhancing customer experience.
Time Series Forecasting Model:
Purpose: Predicts future trends based on historical data.
Example: Forecasting ice cream sales based on seasonal patterns.
Use Cases: Inventory management, budget planning, scheduling.
Anomaly Detection Model:
Purpose: Identifies unusual patterns in data.
Example: Detecting a sudden surge in orders.
Use Cases: Fraud detection, performance monitoring, error identification.
Clustering Model:
Purpose: Groups similar data points together.
Example: Segmenting bookstore customers by preferences.
Use Cases: Targeted marketing, gaining customer insights, resource allocation.
Recipe vs. Model
Recipe:
Definition: A predefined collection of steps and data points.
Purpose: Standardizes workflows and embeds best practices.
Model:
Definition: The output of applying a recipe to data, which represents learned patterns.
Purpose: Analyzes new data and automates decision-making.
8. AI Agents in RapidCanvas
RapidCanvas AI Agents
Definition: Autonomous agents that automate machine learning processes.
Purpose: Simplify the integration and deployment of AI solutions.
Key Features:
Automation
Integration
Intelligence
User-friendliness
Versatility
Example Use Case: Dynamically adjusting product prices in e-commerce.
AskAI: The RapidCanvas AI Agent
Definition: An interactive chatbot within RapidCanvas.
Purpose: Provides insights and answers real-time queries.
How It Works: Processes natural language questions (e.g., “What was our best-selling product last month?”).
9. Schedulers in RapidCanvas
What is a scheduler ?
A scheduler is a defined task or process used to automate workflows and manage data.
Key Features
Automation
Task Execution
Scheduling
Integration
Monitoring and Reporting
Customization
Examples of Schedulers
Data Import Scheduler
Report Generation Scheduler
Alert Scheduler
10. Sharing Your AI Analysis
Methods to Share Outside RapidCanvas
Exporting Reports:
Formats: PDF, Excel/CSV.
Exporting Dashboards:
Formats: Image, PDF.
Exporting Data:
Formats: CSV, Excel.
Using APIs for Data Extraction:
Programmatically extract data.
Exporting AI Models or Solutions:
Formats: PMML, ONNX.
Sharing Visuals:
Capture screenshots or export high-quality images.
Collaborative Sharing:
Share via links or cloud storage.
11. Putting Your AI Solution to Work
Final Steps
Finalize Your Solution:
Review, test, and document the AI model.
Deploy the Model:
Options: On-premises or cloud deployment.
Integration:
Ensure API integration and connect to real-time data sources.
Automate Tasks:
Set up scheduled data processing tasks.
Monitor Performance:
Track key metrics and configure alerts.
User Training and Support:
Educate users and provide necessary help resources.
Continuous Improvement:
Gather feedback and update the model regularly.
Reporting and Analysis:
Summarize outcomes and use insights to inform decisions.
Last updated
