Muon2

Thanks for your advice. Here is the updated setting: ```python optimizer = th.LevenbergMarquardt( objective, linear_solver_cls=th.CholeskyDenseSolver, linearization_cls=th.DenseLinearization, linear_solver_kwargs={'check_singular': False}, vectorize=True, max_iterations=1000, step_size=0.01, abs_err_tolerance = 1e-12, rel_err_tolerance = 1e-10, ) ... with...

> As mentioned above, please keep in mind that we haven't yet added support for line search methods, so we only have basic control over the step sizes taken by...

@luisenp, sorry for the late reply. I tried multiple combinations of hyperparameters, but it still requires a lot of time and does not converge to an acceptable precision. I will...

> > @pixeli99 Thanks for your work. I have a similar question, why you chose the `rand_cosine_interpolated` noise scheduler instead of the one mentioned in the `EDM (Karras et al.)`...

> I understand what you're saying, but I think that different sigma distributions correspond to different diffusion paths. In theory, would it be possible to use the same sampler for...