DashCanvas.us: Fastest way to build data science apps and dashboards
Convert ML Models to Data science apps in a few hours with Low code drag-n-drop app builder that generates Python web apps
Although there are many open source python libraries like Plotly, Dash, Streamlit, Gradio or OpenCV to help data scientists, there is no low-code app builder designed for data scientists to accelerate the development of data science apps. Data scientists are good at building models, however to bring their models to the consumers they have to wait for frontend and backend engineers. The iteration of data science apps takes forever. DashCanvas helps data scientists build and iterate their dashboards orders of magnitude faster.
DashCanvas is designed for data scientists and AI engineers to help them create powerful applications using Dash (Plotly.com), and OpenCV. The low-code drag-n-drop UI builder works from within Jupyter notebook. Once apps are ready, the data science team can distribute them to any number of customers and configure for each customer with a custom configuration. Post distribution, they also can monitor performance and collect feedback from the users of their AI. Feedback can be then used to retrain the ML models.
*****
Data scientist to full stack data scientist in a few hours
DashCanvas helps any data scientist to bridge the gap between a data scientist and a full stack data scientist.
Build Data-science dashboards and apps
There are many verticals where data science dashboards are used. Insurance tech, recommendation engines, ETL analysis, UAV software, security and surveillance, manufacturing supervision or marketing analytics are just some examples. Regardless the use-case, accelerating the development reduces time and cost, key factors in any project. Here is how.
Drag and drop UI builder with Jupyter notebook (optional)
Using a web based drag-n-drop UI designer, data-scientists can build their own apps using Dash or OpenCV. The app’s source code is generated in Python using Dash (plotly.com). The app can be connected with the inference engine using rest API connectors.
*****
Distribute Dashboards & Apps
Developers can add their customers to the system. They can distribute their apps to the customers. A customer can have any number of users. Customers can assign to their users different roles according to their needs like admin, manager, inspector etc. All roles and permissions can be created and modified in the web app.
*****
Edge Fleet Management
When customers of App developers log in to their accounts, they can manage their edge device instances.
Edge case discovery
DashCanvas provides a feedback widget that allows App developers to gather feedback from the users of their AI apps. Data science teams can analyzes that feedback, and fine-tune their ML models.
Performance Analytics
A performance analytics dashboard provides analysis of ML models and apps in production.
Despite being an early-stage product, these features are enabling building, deployment of data science dashboards and apps 3x faster. And it is just the beginning.
*****
MLOps Lifecycle
DashCanvas operates in the deployment section of the MLOps lifecycle. DashCanvas helps AI engineers in the deployment of their AI Apps and collecting feedback from the users of their AI.
MLOps Lifecycle by NVIDIA
As shown in MLOps pipeline here, DashCanvas helps AI engineers to deploy ML systems 5x faster.
ML Environments
As shown in ML environment provided by Gartner, DashCanvas allows AI engineers to deploy, monitor and retrain the models.
ML Data Architecture
For enterprise data pipeline architecture as shown below, application deployment is where DashCanvas helps AI engineers.
