new linter: ruff

.github/workflows/lint.yml

@@ -49,22 +49,5 @@ jobs:
           poetry run pre-commit install
           poetry run pre-commit run --all-files
 
-      - name: Run black
-        run: |
-          poetry run black --check btk
-
-      - name: Run isort
-        run: |
-          poetry run isort --check btk
-
-      - name: Run pydocstyle
-        run: |
-          poetry run pydocstyle btk
-
-      - name: Run flake8
-        run: |
-          poetry run flake8 btk
-
-      - name: Run pylint
-        run: |
-          poetry run pylint btk
+      - name: Run Ruff
+        run: poetry run ruff check --output-format=github btk/ tests/

.pre-commit-config.yaml

@@ -1,7 +1,7 @@
 ---
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.4.0
+    rev: v4.5.0
     hooks:
       - id: trailing-whitespace
         exclude: ^.*fits
@@ -13,8 +13,9 @@ repos:
       - id: check-merge-conflict
       - id: end-of-file-fixer
         exclude: ^.*fits
+
   - repo: https://github.com/kynan/nbstripout
-    rev: 0.6.1
+    rev: 0.7.1
     hooks:
       - id: nbstripout
         args:

btk/blend_batch.py

@@ -432,16 +432,10 @@ def load(cls, path: str, batch_number: int = 0):
                 catalog_list.append(read_table_hdf5(f, path=f"catalog_list/{ii}"))
 
             # load segmentation
-            if "segmentation" in f.keys():
-                segmentation = f["segmentation"][:]
-            else:
-                segmentation = None
+            segmentation = f["segmentation"][:] if "segmentation" in f.keys() else None
 
             # load deblended images
-            if "deblended_images" in f.keys():
-                deblended_images = f["deblended_images"][:]
-            else:
-                deblended_images = None
+            deblended_images = f["deblended_images"][:] if "deblended_images" in f.keys() else None
 
             # load general info about blend
             batch_size = f.attrs["batch_size"]

btk/deblend.py

@@ -14,7 +14,12 @@
 from numpy.linalg import LinAlgError
 from skimage.feature import peak_local_max
 
-from btk.blend_batch import BlendBatch, DeblendBatch, DeblendExample, MultiResolutionBlendBatch
+from btk.blend_batch import (
+    BlendBatch,
+    DeblendBatch,
+    DeblendExample,
+    MultiResolutionBlendBatch,
+)
 from btk.draw_blends import DrawBlendsGenerator
 from btk.multiprocess import multiprocess
 
@@ -39,6 +44,8 @@
         Args:
             ii: The index of the example in the batch.
             blend_batch: Instance of `BlendBatch` class.
+            njobs: Number of processes to use.
+            kwargs: Additional arguments to pass to deblender call.
 
         Returns:
             Instance of `DeblendedExample` class.
@@ -72,7 +79,8 @@
 
         Args:
             blend_batch: Instance of `BlendBatch` class
-            njobs: Number of njobs to paralelize across
+            njobs: Number of jobs to paralelize across
+            kwargs: Additional keyword arguments to pass to each deblend call.
 
         Returns:
             Instance of `DeblendedBatch` class
@@ -126,6 +134,7 @@
         Args:
             ii: The index of the example in the batch.
             mr_batch: Instance of `MultiResolutionBlendBatch` class
+            njobs: Number of processes to use.
 
         Returns:
             Instance of `DeblendedExample` class
@@ -211,10 +220,7 @@
     def deblend(self, ii: int, blend_batch: BlendBatch) -> DeblendExample:
         """Performs measurement on the ii-th example from the batch."""
         blend_image = blend_batch.blend_images[ii]
-        if self.use_mean:
-            image = np.mean(blend_image, axis=0)
-        else:
-            image = blend_image[self.use_band]
+        image = np.mean(blend_image, axis=0) if self.use_mean else blend_image[self.use_band]
 
         # compute threshold value
         threshold = self.threshold_scale * np.std(image)
@@ -281,15 +287,16 @@
         """Performs measurement on the i-th example from the batch."""
         # get a 1-channel input for sep
         blend_image = blend_batch.blend_images[ii]
-        if self.use_mean:
-            image = np.mean(blend_image, axis=0)
-        else:
-            image = blend_image[self.use_band]
+        image = np.mean(blend_image, axis=0) if self.use_mean else blend_image[self.use_band]
 
         # run source extractor
         bkg = sep.Background(image)
         catalog, segmentation = sep.extract(
-            image, self.thresh, err=bkg.globalrms, segmentation_map=True, minarea=self.min_area
+            image,
+            self.thresh,
+            err=bkg.globalrms,
+            segmentation_map=True,
+            minarea=self.min_area,
         )
 
         segmentation_exp = np.zeros((self.max_n_sources, *image.shape), dtype=bool)
@@ -380,10 +387,16 @@
 
                 # add new predictions, masking those that are closer than threshold
                 ra_coordinates = np.concatenate(
-                    [ra_coordinates, ra_detections[distance2d > self.matching_threshold]]
+                    [
+                        ra_coordinates,
+                        ra_detections[distance2d > self.matching_threshold],
+                    ]
                 )
                 dec_coordinates = np.concatenate(
-                    [dec_coordinates, dec_detections[distance2d > self.matching_threshold]]
+                    [
+                        dec_coordinates,
+                        dec_detections[distance2d > self.matching_threshold],
+                    ]
                 )
             else:
                 ra_coordinates = np.concatenate([ra_coordinates, ra_detections])
@@ -441,7 +454,7 @@
         Args:
             ii: The index of the example in the batch.
             blend_batch: Instance of `BlendBatch` class.
-            reference_catalog: Reference catalog to use for deblending. If None, the
+            reference_catalogs: Reference catalog to use for deblending. If None, the
                 truth catalog is used.
 
         Returns:

btk/draw_blends.py

@@ -128,8 +128,7 @@ def get_catsim_galaxy(
         agn = galsim.Gaussian(flux=agn_flux, sigma=1e-8)
         components.append(agn)
 
-    profile = galsim.Add(components)
-    return profile
+    return galsim.Add(components)
 
 
 class DrawBlendsGenerator(ABC):

btk/match.py

@@ -195,6 +195,7 @@ def __init__(self, pixel_max_sep=5.0, **kwargs) -> None:
 
         Args:
             pixel_max_sep: the maximum separation in pixels to be considered a match
+            kwargs: Additional arguments for parent class.
         """
         super().__init__(**kwargs)
         self.distance_function = pixel_l2_distance_matrix
@@ -236,6 +237,7 @@ def __init__(self, arcsec_max_sep=2.0, **kwargs) -> None:
 
         Args:
             arcsec_max_sep: the maximum separation in arcsec to be considered a match
+            kwargs: Additional arguments for parent class.
         """
         super().__init__(**kwargs)
         self.max_sep = arcsec_max_sep

btk/measure.py

@@ -42,7 +42,7 @@
 ) -> np.ndarray:
     """Return ellipticities of both true and detected galaxies, assuming they are matched."""
     # psf is assumed to be the same for the entire batch and correspond to selected band.
-    assert len(images.shape) == 4  # (batch_size, max_n_sources, H, W)
+    assert images.ndim == 4  # (batch_size, max_n_sources, H, W)
     batch_size, max_n_sources, _, _ = images.shape
     ellipticities = np.zeros((batch_size, max_n_sources, 2))
     for ii in range(batch_size):

btk/metrics/reconstruction.py

@@ -27,8 +27,8 @@ def _get_recon_metric(
                 images2 = iso_images2[ii, :n_sources]
                 mets = metric_func(images1, images2)
                 assert mets.shape == (n_sources,)
-                for ii in range(n_sources):
-                    results.append(mets[ii])
+                for jj in range(n_sources):
+                    results.append(mets[jj])
 
         return np.array(results)
 

btk/metrics/utils.py

@@ -27,7 +27,7 @@
     return np.where(is_on, seg, np.nan)
 
 
-def mse(image1: np.ndarray, image2: np.ndarray) -> np.ndarray:
+def mse(images1: np.ndarray, images2: np.ndarray) -> np.ndarray:
     """Computes mean-squared-error (MSE) between two images.
 
     Args:
@@ -39,7 +39,7 @@
     Returns:
         Returns MSE between each corresponding iamge as an array of shape `*`.
     """
-    return np.sqrt(np.power((image1 - image2), 2).mean(axis=(-1, -2)))
+    return np.sqrt(np.power((images1 - images2), 2).mean(axis=(-1, -2)))
 
 
 def iou(seg1: np.ndarray, seg2: np.ndarray) -> np.ndarray:
@@ -65,7 +65,7 @@
     return i / u
 
 
-def psnr(image1: np.ndarray, image2: np.ndarray) -> np.ndarray:
+def psnr(images1: np.ndarray, images2: np.ndarray) -> np.ndarray:
     """Compute peak-signal-to-noise-ratio from skimage.
 
     Args:
@@ -77,18 +77,18 @@
     Returns:
         Returns PSNR between each corresponding iamge as an array of shape `N`.
     """
-    n, h, w = image1.shape
-    assert image1.min() >= 0 and image2.min() >= 0
-    assert (n, h, w) == image2.shape
+    n, h, w = images1.shape
+    assert images1.min() >= 0 and images2.min() >= 0
+    assert (n, h, w) == images2.shape
     psnr_ = np.zeros(n)
     for ii in range(n):
-        im1 = image1[ii] / image1[ii].max()
-        im2 = image2[ii] / image2[ii].max()
+        im1 = images1[ii] / images1[ii].max()
+        im2 = images2[ii] / images2[ii].max()
         psnr_[ii] = peak_signal_noise_ratio(im1, im2, data_range=1)
     return psnr_
 
 
-def struct_sim(image1: np.ndarray, image2: np.ndarray, **kwargs) -> np.ndarray:
+def struct_sim(images1: np.ndarray, images2: np.ndarray, **kwargs) -> np.ndarray:
     """Compute structural similarity index from skimage.
 
     By default, we normalize the images to be between 0 and 1. So that the
@@ -106,13 +106,13 @@
         Returns structural similarity index between each corresponding iamge as
         an array of shape `N`.
     """
-    assert image1.min() >= 0 and image2.min() >= 0
-    n, h, w = image1.shape
-    assert (n, h, w) == image2.shape
+    assert images1.min() >= 0 and images2.min() >= 0
+    n, h, w = images1.shape
+    assert (n, h, w) == images2.shape
     ssim = np.zeros(n)
     for ii in range(n):
-        im1 = image1[ii] / image1[ii].max()
-        im2 = image2[ii] / image2[ii].max()
+        im1 = images1[ii] / images1[ii].max()
+        im2 = images2[ii] / images2[ii].max()
         ssim[ii] = structural_similarity(im1, im2, data_range=1, **kwargs)
     return ssim
 

btk/multiprocess.py

@@ -2,7 +2,7 @@
 
 import multiprocessing as mp
 from itertools import repeat, starmap
-from typing import Callable, Iterable
+from typing import Callable, Iterable, Optional
 
 
 def _apply_args_and_kwargs(func: Callable, args, kwargs):
@@ -29,7 +29,7 @@ def get_current_process() -> int:
 def multiprocess(
     func: Callable,
     args_iter: Iterable,
-    kwargs_iter: Iterable = None,
+    kwargs_iter: Optional[Iterable] = None,
     njobs: int = 1,
     verbose: bool = False,
 ):

btk/sampling_functions.py

@@ -194,7 +194,7 @@ def __init__(
         self.stamp_size = stamp_size
         self.min_mag, self.max_mag = min_mag, max_mag
         self.mag_name = mag_name
-        self.max_shift = self.stamp_size / 2 if not max_shift else max_shift
+        self.max_shift = max_shift if max_shift else self.stamp_size / 2
 
         # only within area where sources are allowed
         self.exp_count = density * (self.max_shift * 2 / 60) ** 2

btk/survey.py

@@ -74,7 +74,7 @@
 
 
 def get_surveys(
-    names: Union[str, List[str]], psf_func: Callable = None
+    names: Union[str, List[str]], psf_func: Optional[Callable] = None
 ) -> Union[Survey, List[Survey]]:
     """Return specified surveys from galcheat extended to contain PSF information.
 
@@ -205,12 +205,11 @@
     else:
         raise RuntimeError(f"No psf files found in '{psf_dir}'.")
     psf_array = fits.getdata(psf_dir + "/" + psf_file)
-    psf_model = galsim.InterpolatedImage(
+
+    return galsim.InterpolatedImage(
         galsim.Image(psf_array), scale=survey.pixel_scale.to_value("arcsec")
     ).withFlux(1.0)
 
-    return psf_model
-
 
 def make_wcs(
     pixel_scale: float,