Skip to content

Precisionboost > 2 tends to break the PowerSpectrum calculation #28

Description

@JulianBMunoz

The Coefficients and UVLF work fine, but the arrays get TOO BIG for PowerSpectrum, given the new changes with vectorization.

Suggested solution: create new precisionboost parameters separate for z, r, R, and others if needed. Keep the precisionboost too as a bruteforce way to increase all.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Labels

No labels
No labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions