dads-thesis
casadi
| dads-thesis | casadi | |
|---|---|---|
| 1 | 4 | |
| 0 | 1,554 | |
| - | 1.7% | |
| 0.0 | 9.3 | |
| almost 4 years ago | 2 days ago | |
| Fortran | C++ | |
| - | GNU Lesser General Public License v3.0 only |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
dads-thesis
-
Things that makes your jaw drop open and not know what to say...
I have a standing project to try and get the code from my dad's master's thesis working again. A lot of fortran code is used for numerical computing, so the APIs are a lot simpler than some sort of industrial control software. Now, Fortran IV is old as hell (F66 and F77 have command line flags for gfortran, but I'm not sure about earlier versions) but at least there is a lot of tooling for you here.
casadi
-
pyomo VS casadi - a user suggested alternative
2 projects | 5 Sep 2023
Interface for several solvers and integrators.
-
(Direct) Collocation in (Time) Optimal Control
Howdy! Collocation methods can be... tricky. For NMPC control of vehicles, success has been had using direct multiple shooting. Also easier to implement and more intuitive. In fact, this example from the GH is pretty instructive: https://github.com/casadi/casadi/blob/master/docs/examples/python/race_car.py
-
Are there any optimization libraries/packages that use automatic differentiation?
JuMP.jl (Julia) or casADi (Python) are good choices.
-
Should I switch over completely to Julia from Python for numerical analysis/computing?
Python is not mature in this area. If you ask Google what Simulink for Python is, you get responses that point to dead libraries that were never feature complete and slow. The absolute closest is CASADI which is nice for some things but doesn't even have a true causal modeling interface and is mostly abandoned by the developers (they put a patch in every now and then, but just look at the commit graph), and it's slow compared to the Julia tools, so much so that PyBAMM is interfacing with ModelingToolkit.jl in Julia for a performance boost. Python is not the place to be for causal/acausal modeling or controls. Anyone who is saying "Python is mature" here is saying it in the abstract and not in the context of your actual question. Yes, Python has web development frameworks. No it does not have good libraries for tons of areas (control, acausal modeling, pharmacometrics, etc.).
What are some alternatives?
ceres-solver - A large scale non-linear optimization library
ModelingToolkit.jl - An acausal modeling framework for automatically parallelized scientific machine learning (SciML) in Julia. A computer algebra system for integrated symbolics for physics-informed machine learning and automated transformations of differential equations
symbolic - A Symbolic Package for Octave using SymPy
jsbsim - An open source flight dynamics & control software library
symforce - Fast symbolic computation, code generation, and nonlinear optimization for robotics
fricas - Official repository of the FriCAS computer algebra system
wyvern - Automatic conversion of call by value into call by need in the LLVM IR.
Causal.jl - Causal.jl - A modeling and simulation framework adopting causal modeling approach.
NumCpp - C++ implementation of the Python Numpy library
pyomo - An object-oriented algebraic modeling language in Python for structured optimization problems.
Catalyst.jl - Chemical reaction network and systems biology interface for scientific machine learning (SciML). High performance, GPU-parallelized, and O(1) solvers in open source software.
MomentClosure.jl - Tools to generate and study moment equations for any chemical reaction network using various moment closure approximations