Add support for TOF imaging #309
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## next #309 +/- ##
==========================================
- Coverage 94.62% 93.73% -0.90%
==========================================
Files 21 21
Lines 1266 1277 +11
==========================================
- Hits 1198 1197 -1
- Misses 68 80 +12 ☔ View full report in Codecov by Sentry. |
KedoKudo
changed the title
| if all(matches): | ||
| logger.info("Using rotation angles from filenames.") | ||
| rotation_angles = np.array([float(".".join(m.groups())) for m in matches]) | ||
| regex = r"\d{8}_\S*_\d{4}_(?P<deg>\d{3})_(?P<dec>\d{3})_\d*\.(?:tiff?|fits)" |
There was a problem hiding this comment.
You should add tests for this function to press on this regex a lot harder
| @@ -103,6 +103,7 @@ def calculate_dissimilarity( | |||
| tilt: float, | |||
| image0: np.ndarray, | |||
| image1: np.ndarray, | |||
| center: Optional[Tuple[Union[float, int], Union[float, int]]] = None, | |||
There was a problem hiding this comment.
This would be more readable if you created a custom type. Maybe
type Number = Union[float, int]
type Pair = Tuple[Number, Number]
...
center: Optional[Pair] = None,There was a problem hiding this comment.
it will be easier once we make 3.11 the minimum version of python as we can write it as Optional[Tuple[float|int, float|int]].
There was a problem hiding this comment.
I think that the compound typing you have is harder to read. You can keep it how you want, but center: Optional[Tuple[Number, Number]] = None is significantly clearer to me.
| @@ -124,6 +124,24 @@ def test_normalization_bright_dark(): | |||
| assert diff < 0.01 | |||
|
|
|||
|
|
|||
| def test_normalization_no_darks(): | |||
There was a problem hiding this comment.
Do you know about numpy.testing?
There was a problem hiding this comment.
I am assuming you are asking to use np.testing.assert_all_close instead of computing the average difference here. They are similar in function, but I prefer the explicit one here over the one provided by numpy
There was a problem hiding this comment.
Just a thing to point out as existing
provide file type in error messages as suggested Co-authored-by: Pete Peterson <[email protected]>
use consistent type hinting Co-authored-by: Pete Peterson <[email protected]>
Just the type annotations. We are not using mypy for iMars3D. |
This PR introduces this following changes:
memmap(the instrument team reported that it is slowing down the reading).tiffand.tifare considered valid extension for TIFF input format nowdarks=Noneto assume perfect dark fields (all dark fields are zero)Noneif extraction fails. For files with partially missing rotation angles, the loader will returnnp.nanin the rotation angle list, allowing users to handle or sanitize the missing values as needed. While files must still adhere to the correct naming pattern (or contain the necessary metadata) for auto reduction, the auto reconstruction process will no longer fail if rotation angle extraction is unsuccessful.