juliaturkudatascience / fdesolver.jl Goto Github PK

The default licence when you create a template package is MIT, but we can also go for another one. For now, the licence is stored in this file.

You used .= in some parts of the main code. For example, in line 24 we have: Y[2, :] .= T0 .+ h .^ β .* F(t, n, β, Y, par...) ./ Γ(β .+ 1)

I found it problematic in some cases, e.g. for the solution of Gonze model with N species.

So, I replaced = instead of .= from the main.jl.

If you think the presence of .= is so crucial, let me know to think more about it. If not, let's remove dots.

This issue is used to trigger TagBot; feel free to unsubscribe.

If you haven't already, you should update your TagBot.yml to include issue comment triggers.
Please see this post on Discourse for instructions and more details.

If you'd like for me to do this for you, comment TagBot fix on this issue.
I'll open a PR within a few hours, please be patient!

@JuliaRegistrator register

Is there a possibility to have an (automated) package homepage, similar to R pkgdown. See for instance mia homepage

@JuliaRegistrator register

Before releasing the solver with FFT, we need to add some tests on this newer version. We can update the older tests or make new ones.

@JuliaRegistrator register

@JuliaRegistrator register

Travis CI enables continuous integration of a GitHub repository and codecov check that the code is tested to avoid errors and exceptions. They are free for public repositories, so we can set them up after we decide to make our repo public. They can also be run on shared repositories belonging to an organisation / team.

• To avoid duplication, keep examples in a single place and link there from README. Do not duplicate.
• Add relevant references from the authors in README and examples (I already added some)

Just wanted to let you know there is a minor error in the first example of the Readme. Running it now gives:

ERROR: LoadError: MethodError: no method matching F(::Float64, ::Vector{Float64}) Closest candidates are: F(::Any, ::Any, ::Any) at ...

As par seems to be missing from the FDEsolver call.

Otherwise, it was much fun to play around with the package, great work!

[pos_args.tSpan[2] - t(N)] should be changed to [pos_args.tSpan[2] - t[N]]

Here we are going to release v1.0.0 of FdeSolver.jl, which now solves fractional differential equations in a faster way thanks to Fast-Fourier Transform (FFT).

In API

We bring examples here. I will edit the examples and choose which one should be there.
and delete the Examples file instead.
We don't need any extra documentation else than README

@JuliaRegistrator register

The current README relies on copypasted figures. Is it possible to automated figure generation. With Julia for Rmarkdown or something similar, Julia specific.

We should fix warnings for tol, itmax, and nc.

tol and nc are for the explicit method, but tol and itmax are for the implicit method (with Jacobian).
We should also make separate the itmax and nc in the method without memory.
Thanks to using the dispatch feature, we are not worried about using an argument that is not related to the method.

We need to compare execution time of our solver against its MATLAB counterpart.

Current challenges:

• interpolate MATLAB solvers in Julia
• make sure that interpolation does not affect execution time
• benchmark and compare Julia and MATLAB results in terms of accuracy and efficiency

Finally, we can register our new package using the app [Registrator]https://awesomeopensource.com/project/JuliaRegistries/Registrator.jl) and open an issue where we tell the registrator bot @JuliaRegistrator register() and the package will be elected for the register.

@JuliaRegistrator register

This issue can be used when the second release of the package is ready. New features should be listed:

• Jacobian case
• Convergence / Tolerance case
• Published documentation
• ...

When we update the README.md in the repository, the changes are not reflected in the GitHub Pages documentation.

I try to solve it with GHA. The GitHub Actions workflow for documentation (documentation.yml) runs successfully without errors. Despite this, the updates to the README aren't appearing in the stable version of the documentation. The issue persists despite ensuring the correct setup in make.jl and the workflow file.

[pos_args.tSpan[2] - t(N)] -> [pos_args.tSpan[2] - t[N]]