Tilman Troester

camb.model.*Params attributes default to zero

3

When creating a new instance of one of the `camb.model.*Params` classes, the attributes are set to zero by default instead of the sensible defaults in `model.f90`. For example, in case...

I found that running zeus with n_MPI = n_walker/2 gives poor efficiency, with half of the CPU time being spend idling. If I understand the code in `EnsembleSampler.sample` correctly, the...

This PR enables CPU-only builds on macOS. It in addition to the changes to `spconv`, this also requires some changes to `ccimport` and `cumm`. They are implemented at https://github.com/tilmantroester/ccimport/tree/macos_support and...

`write_yaml` does not serialise completely

3

`write_yaml` currently ignores the `baryonic_effects` and `mg_parametrization` arguments to `Cosmology`, since it only uses what's in `self._params_init_kwargs` and `self._config_init_kwargs`. It also loses the information on `transfer_function` and `matter_power_spectrum` if they...

`CCLNamedClass` does not play nice with `dataclass`

3

I tried making a `dataclass` that derived from `CCLNamedClass` (analogously to the `Baryons` base class) but this causes an obscure error: ```python from dataclasses import dataclass from pyccl import CCLNamedClass...

BaccoemuBaryons does not check for k or a ranges

3

I just tried `BaccoemuBaryons` but ran into a bunch of issues. The following snipped fails with an opaque error, because the k and a ranges in https://github.com/LSSTDESC/CCL/blob/master/pyccl/baryons/baccoemu_baryons.py#L47 are never checked:...

I like OOP but playing around with the new API I found that it doesn't play nice with editors. In the example ```python cosmo = ccl.CosmologyVanillaLCDM() cosmo.angular_cl ``` VS Code...

Proposal for likelihood specification and factory API

1

This is a more-or-less working prototype of a yaml-based likelihood specification, following the [discussion thread](https://github.com/LSSTDESC/firecrown/discussions/308#discussioncomment-7589995). Some features I think are nice: - Separation of model and data specification: specifying the...

API for passing sacc data to `load_likelihood` and `build_likelihood`

3

For blinding (@arthurmloureiro) we need to be able to pass sacc objects to the likelihood creation entry point. At the moment, `load_likelihood` and `build_likelihood` expect a `NamedParameters` argument. This precludes...

Add a sacc-less utility "likelihood"

1

Right now the likelihoods in firecrown require a sacc object to compute anything. It would be nice to have a utility likelihood that takes in a list of ell (or...