# NeuralODE and ModelingToolkit
**URL:** https://discourse.julialang.org/t/neuralode-and-modelingtoolkit/43855
**Category:** Machine Learning
**Tags:** diffeq, sciml
**Created:** [July 29, 2020, 3:08am UTC](https://discourse.julialang.org/t/neuralode-and-modelingtoolkit/43855 "2020-07-29T03:08:39Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@Jagirhussan](https://discourse.julialang.org/u/Jagirhussan)
#### Post date: [July 29, 2020, 3:08am UTC](https://discourse.julialang.org/t/neuralode-and-modelingtoolkit/43855/1 "2020-07-29T03:08:39Z")
I would like to do statespace analysis. ModelingToolkit provides some very useful constructs to do this. I was wondering if there is a way to analyse NeuralODE’s (trained using DiffEqFlux) using ModelingToolkit?
Thanks
---
### Author:  [@ChrisRackauckas](https://discourse.julialang.org/u/ChrisRackauckas)
#### Post date: [July 29, 2020, 3:50am UTC](https://discourse.julialang.org/t/neuralode-and-modelingtoolkit/43855/2 "2020-07-29T03:50:08Z")
[DataDrivenDiffEq.jl](https://datadriven.sciml.ai/dev/) has a nice way of transforming neural ODEs back to MTK symbolic expressions via [SInDy](https://datadriven.sciml.ai/dev/sparse_identification/sindy/). I explain exactly how to do this in the [scientific machine learning workshop](https://www.youtube.com/watch?v=QwVO0Xh2Hbg) (like 3 hours in?)
---
### Author:  [@Jagirhussan](https://discourse.julialang.org/u/Jagirhussan)
#### Post date: [July 29, 2020, 5:07am UTC](https://discourse.julialang.org/t/neuralode-and-modelingtoolkit/43855/3 "2020-07-29T05:07:31Z")
Many thanks for the prompt response. It is at 3hr 16 mins in the youtube video.
I have a reaction, will try ISInDy and see how it goes.