Get model components with current parameters guess

[DSimonne]

Hello

Is it possible to get the values over a range x of the components of a model, before fitting, with a set of guess parameters ?
I try to explain:
When I create a model, I can get an initial guess by typing:

init = mod.eval(pars, x=x)

But to get the components values, I can only type:

components = mod.eval_components(x=x)

without giving parameters. And the resulting components are computed with the basic parameters for each component (e.g. amplitude = 1, etc ...).

Of course then I can fit the model and get the components with:

out = mod.fit()
components = out.eval_components(x=x)

But I would like to plot my initial guess before fitting to see how accurate it is.

Thank you for the help, I hope it is clear !

David

[newville]

@DSimonne yeah, I guess that CompositeModel.eval_components should have and use a params keyword arg.

A PR would be welcome.

[DSimonne]

Hi @newville,

I'll try and work on that !