Template based code generation for networks of hybrid systems - - PowerPoint PPT Presentation

▶
template based code generation for networks of hybrid
SMART_READER_LITE
LIVE PREVIEW

Template based code generation for networks of hybrid systems - - PowerPoint PPT Presentation

Sep 16, 2023 182 likes •480 views

Template based code generation for networks of hybrid systems Cambdridge CodeGen Workshop Boris Marin Silver Lab University College London January 7, 2014 Boris Marin (University College London) 1 / 7 Model specifiation languages

slide-1
SLIDE 1

Template based code generation for networks of hybrid systems

Cambdridge CodeGen Workshop Boris Marin

Silver Lab University College London

January 7, 2014

Boris Marin (University College London) 1 / 7

slide-2
SLIDE 2

Model specifiation languages

Requirements

(in the context of Computational Neuroscience)

discrete units, modeled as dynamical systems

ODEs / maps SDEs / kinetic schemes

discontinuous state jumps (e.g. synaptic coupling)

event detection (rootfinding)

networks of units

event propagation

In summary, networks of hybrid systems

Boris Marin (University College London) 2 / 7

slide-3
SLIDE 3

Model specifiation languages

Requirements

(in the context of Computational Neuroscience)

discrete units, modeled as dynamical systems

ODEs / maps SDEs / kinetic schemes

discontinuous state jumps (e.g. synaptic coupling)

event detection (rootfinding)

networks of units

event propagation

In summary, networks of hybrid systems

Boris Marin (University College London) 2 / 7

slide-4
SLIDE 4

Model specifiation languages

Requirements

(in the context of Computational Neuroscience)

discrete units, modeled as dynamical systems

ODEs / maps SDEs / kinetic schemes

discontinuous state jumps (e.g. synaptic coupling)

event detection (rootfinding)

networks of units

event propagation

In summary, networks of hybrid systems

Boris Marin (University College London) 2 / 7

slide-5
SLIDE 5

Model specifiation languages

Requirements

(in the context of Computational Neuroscience)

discrete units, modeled as dynamical systems

ODEs / maps SDEs / kinetic schemes

discontinuous state jumps (e.g. synaptic coupling)

event detection (rootfinding)

networks of units

event propagation

In summary, networks of hybrid systems

Boris Marin (University College London) 2 / 7

slide-6
SLIDE 6

Model specifiation languages

Requirements

(in the context of Computational Neuroscience)

discrete units, modeled as dynamical systems

ODEs / maps SDEs / kinetic schemes

discontinuous state jumps (e.g. synaptic coupling)

event detection (rootfinding)

networks of units

event propagation

In summary, networks of hybrid systems

Boris Marin (University College London) 2 / 7

slide-7
SLIDE 7

Model specifiation languages

Requirements

(in the context of Computational Neuroscience)

discrete units, modeled as dynamical systems

ODEs / maps SDEs / kinetic schemes

discontinuous state jumps (e.g. synaptic coupling)

event detection (rootfinding)

networks of units

event propagation

In summary, networks of hybrid systems

Boris Marin (University College London) 2 / 7

slide-8
SLIDE 8

Model specifiation languages

Requirements

(in the context of Computational Neuroscience)

discrete units, modeled as dynamical systems

ODEs / maps SDEs / kinetic schemes

discontinuous state jumps (e.g. synaptic coupling)

event detection (rootfinding)

networks of units

event propagation

In summary, networks of hybrid systems

Boris Marin (University College London) 2 / 7

slide-9
SLIDE 9

Model specifiation languages

Requirements

(in the context of Computational Neuroscience)

discrete units, modeled as dynamical systems

ODEs / maps SDEs / kinetic schemes

discontinuous state jumps (e.g. synaptic coupling)

event detection (rootfinding)

networks of units

event propagation

In summary, networks of hybrid systems

Boris Marin (University College London) 2 / 7

slide-10
SLIDE 10

Model specifiation languages

Requirements

(in the context of Computational Neuroscience)

discrete units, modeled as dynamical systems

ODEs / maps SDEs / kinetic schemes

discontinuous state jumps (e.g. synaptic coupling)

event detection (rootfinding)

networks of units

event propagation

In summary, networks of hybrid systems

Boris Marin (University College London) 2 / 7

slide-11
SLIDE 11

Model specifiation languages

LEMS: a general purpose model specifiation language

SPOILER ALERT!

Boris Marin (University College London) 3 / 7

slide-12
SLIDE 12

Model specifiation languages

LEMS: a general purpose model specifiation language

Boris Marin (University College London) 3 / 7

slide-13
SLIDE 13

Simulating hybrid systems

From MSL to simulation

Bottom line: Simulation of hybrid systems is a well established field. Why not using state of the art numerical libraries? numerical stability

stiffness Zeno

error control (variable stepsize) accurate event detection

Boris Marin (University College London) 4 / 7

slide-14
SLIDE 14

Simulating hybrid systems

From MSL to simulation

Bottom line: Simulation of hybrid systems is a well established field. Why not using state of the art numerical libraries? numerical stability

stiffness Zeno

error control (variable stepsize) accurate event detection

Boris Marin (University College London) 4 / 7

slide-15
SLIDE 15

Simulating hybrid systems

From MSL to simulation

Bottom line: Simulation of hybrid systems is a well established field. Why not using state of the art numerical libraries? numerical stability

stiffness Zeno

error control (variable stepsize) accurate event detection

Boris Marin (University College London) 4 / 7

slide-16
SLIDE 16

Simulating hybrid systems

From MSL to simulation

Bottom line: Simulation of hybrid systems is a well established field. Why not using state of the art numerical libraries? numerical stability

stiffness Zeno

error control (variable stepsize) accurate event detection

Boris Marin (University College London) 4 / 7

slide-17
SLIDE 17

Simulating hybrid systems

From MSL to simulation

Bottom line: Simulation of hybrid systems is a well established field. Why not using state of the art numerical libraries? numerical stability

stiffness Zeno

error control (variable stepsize) accurate event detection

Boris Marin (University College London) 4 / 7

slide-18
SLIDE 18

Our contribution

The need for an intermediate format

template-based code generation: attract expert users

close correspondence to target structure most targets use a similar format

alternative approach: develop a compiler

Boris Marin (University College London) 5 / 7

slide-19
SLIDE 19

Our contribution

The need for an intermediate format

template-based code generation: attract expert users

close correspondence to target structure most targets use a similar format

alternative approach: develop a compiler

Boris Marin (University College London) 5 / 7

slide-20
SLIDE 20

Our contribution

The need for an intermediate format

template-based code generation: attract expert users

close correspondence to target structure most targets use a similar format

alternative approach: develop a compiler

Boris Marin (University College London) 5 / 7

slide-21
SLIDE 21

Our contribution

The need for an intermediate format

template-based code generation: attract expert users

close correspondence to target structure most targets use a similar format

alternative approach: develop a compiler

Boris Marin (University College London) 5 / 7

slide-22
SLIDE 22

Our contribution

The distilled LEMS format

Goal: direct mapping to widely used ODE steppers with event handling

Boris Marin (University College London) 6 / 7

slide-23
SLIDE 23

Our contribution

The distilled LEMS format

Boris Marin (University College London) 6 / 7

slide-24
SLIDE 24

Our contribution

Current status

LEMS → dLEMS → LLNL Sundials (CVODE/IDA)

matlab ODEsuite

XPPAUT

modelica (nearly) effortless accomodation of addditional formats (nodes only: no event routing)

Boris Marin (University College London) 7 / 7

slide-25
SLIDE 25

Our contribution

Current status

LEMS → dLEMS → LLNL Sundials (CVODE/IDA)

matlab ODEsuite

XPPAUT

modelica (nearly) effortless accomodation of addditional formats (nodes only: no event routing)

Boris Marin (University College London) 7 / 7

slide-26
SLIDE 26

Our contribution

Current status

LEMS → dLEMS → LLNL Sundials (CVODE/IDA)

matlab ODEsuite

XPPAUT

modelica (nearly) effortless accomodation of addditional formats (nodes only: no event routing)

Boris Marin (University College London) 7 / 7

slide-27
SLIDE 27

Our contribution

Current status

LEMS → dLEMS → LLNL Sundials (CVODE/IDA)

matlab ODEsuite

XPPAUT

modelica (nearly) effortless accomodation of addditional formats (nodes only: no event routing)

Boris Marin (University College London) 7 / 7

slide-28
SLIDE 28

Our contribution

Current status

LEMS → dLEMS → LLNL Sundials (CVODE/IDA)

matlab ODEsuite

XPPAUT

modelica (nearly) effortless accomodation of addditional formats (nodes only: no event routing)

Boris Marin (University College London) 7 / 7

slide-29
SLIDE 29

Our contribution

Current status

LEMS → dLEMS → LLNL Sundials (CVODE/IDA)

matlab ODEsuite

XPPAUT

modelica (nearly) effortless accomodation of addditional formats (nodes only: no event routing)

Boris Marin (University College London) 7 / 7