Description:

The BlockCanvas project provides a visual environment for creating
simulation experiments, where function and data are separated. Thus,
you can define your simulation algorithm by visually connecting
function blocks into a data flow network, and then run it with various
data sets (known as "contexts"); likewise, you can use the same
context in a different functional simulation.

The project provides support for plotting, function searching and
inspection, and optimization. It includes a stand-alone application
that demonstrates the block-canvas environment, but the same
functionality can be incorporated into other applications.