kmolan / multicalc-rust Goto Github PK

Thank you for doing this great job!

I am trying to make use of this crate to generate jacobian matrix for curve fitting of arbitrary functions with crate levenberg-marquardt. Therefore, I need to derive jacobian matrix based on the generated closures that capture input xdata in runtime. Do you have any plan to support closure as the element of function_matrix in the module of Jacabian?

I am using this crate to do curve fitting with 'nalgebra' that is an algebra crate. For all I know, most algebra library tends to use column-major matrix and store matrix in a vec storage. Hence, I think it is better to provide new api or pass the paramter to control the output matrix shape to be:

• row-major: [[T; ncols]; nrows]
• column-major: [[T; nrows]; ncols]
• contiguous column by column: [T; nrows * ncols] => default for many crates
• contiguous row by row: [T; nrows * ncols]

Thank you for your great job. It works very well in my crate right now.

The current default step is pub const DEFAULT_STEP_SIZE: f64 = 0.00001;. It is too big for some functions. I would recommend a smaller value pub const DEFAULT_STEP_SIZE: f64 = 1.49e-8. You can find the magic value in a popular python library scipy: https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.approx_fprime.html.