Shashi Gowda

WIP Lightweight nodes

5

Nodes now hold exactly the amount of information needed to represent their domain of signals. removed the "Renderer" abstraction in favor of `pull`ing samples from a node. Mostly inlinable `sampleat`...

Take the example of NeuralPDE.jl https://github.com/SciML/NeuralPDE.jl/blob/master/Project.toml it depends on SymbolicUtils, Symbolics, and ModelingToolkit ```julia ModelingToolkit = "6.3.1, 6" SymbolicUtils = "0.13, 0.14, 0.15, 0.16" Symbolics = "3" ``` And Symbolics...

noinline in ShardedForm

4

Fixes #651

Updates for Unityper types

1

must be tested against SU master ┆Issue is synchronized with this [Trello card](https://trello.com/c/1mMVHSR6) by [Unito](https://www.unito.io)

[WIP] Very alpha docs upgrade

4

Made a LaTeX rendering REPL in Franklin and started to write some docs, trying to see if I can deploy it on netlify... What it looks like:  ┆Issue is...

Can verify against ForwardDiff.

We noticed that multi-threaded code where the function is divided into "closures" is wayy faster to compile than a giant serial function. With the trick of interpolating RGFs into RGFs,...

Array differentiation

1

We need this in 3 fronts: 1. Using the tensor notation for operations that can be encoded in that way: http://www.matrixcalculus.org/tensorcalculus.pdf 1. For operations that cannot be encoded in this...

WIP: Array optimization rules

3

Getting this started, let's use this PR for collecting ideas as well. Builds on #495 ┆Issue is synchronized with this [Trello card](https://trello.com/c/b7v4XsIS) by [Unito](https://www.unito.io)

This has been on people's minds for a while but had no issue, https://github.com/JuliaSymbolics/SymbolicUtils.jl/issues/269 https://github.com/JuliaSymbolics/SymbolicUtils.jl/issues/327 Using syntax ChrisRackauckas wrote down in #511 ```julia @variables a::Real # real under the assumption...