Skip to content

Commit

Permalink
comment out JAX
Browse files Browse the repository at this point in the history
  • Loading branch information
Jammy2211 committed Mar 9, 2024
1 parent ca8dc57 commit 4bd6641
Show file tree
Hide file tree
Showing 3 changed files with 2 additions and 47 deletions.
1 change: 0 additions & 1 deletion autoarray/inversion/inversion/abstract.py
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,6 @@
from autoarray.numba_util import profile_func

from autoarray.inversion.linear_obj.linear_obj import LinearObj
from autoarray.inversion.linear_obj.func_list import AbstractLinearObjFuncList
from autoarray.inversion.pixelization.mappers.abstract import AbstractMapper
from autoarray.inversion.regularization.abstract import AbstractRegularization
from autoarray.inversion.inversion.settings import SettingsInversion
Expand Down
4 changes: 2 additions & 2 deletions optional_requirements.txt
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
pylops>=1.10.0,<=1.18.3
pynufft
jax==0.4.3
jaxlib==0.4.3
#jax==0.4.3
#jaxlib==0.4.3
#numba
44 changes: 0 additions & 44 deletions to_do_list
Original file line number Diff line number Diff line change
@@ -1,44 +0,0 @@


- Think very carefully about how Preloads and w_tilde interface with database. Pickle array? Large filesize
but fast. Need to be careful with galaxies.


__Refactor__

- Use composition in visualization MatPlot objects instead of inheritance.
- Extract geometry from mask package.
- auto_labels in visuals as a config loady thing.
- Split wrap / mat_wrap up into separate moduels.

__Visualization__

- Need to properly make it so one can disable many aspections of visualization.


__Inversion With Multiple Multiwavelength Datasets__

Can create a list of InversionMatrix objects for each dataset + tracer and then have a new class which takes a list of
these objects and creates a combined curvature_reg matrix from them, which solves for all datasets simultaneously.

This is genuine because it can reuse all existing functionality and doesn't require diting of the inner code
to get multi dataset fits working.

The regularization matrix could addition addition multi-data regularization terms added.

Multi analysis API can easily be used for model parameterization, with CombineAnalysis overwritten to define
the log likelihood fuinction (which maybe uses a special type of Fit object).



__Probably Terrible Idea Notes__

w_tilde_tlc:

- Faster calculation of w_tilde_imaging (or quick way to get from old one?).

- docstrings in Preloads for w_tilde.

- Proper unit test on Preloads for w_tilde.

- Add options for implciit preload of w_tilde / rest on and off, and manual input of snr_cut? Most likely make it so

0 comments on commit 4bd6641

Please sign in to comment.