use logger

giga_connectome/atlas.py

@@ -9,6 +9,10 @@
 from nibabel import Nifti1Image
 from pkg_resources import resource_filename
 
+from giga_connectome.logger import gc_logger
+
+gc_log = gc_logger()
+
 
 PRESET_ATLAS = ["DiFuMo", "MIST", "Schaefer20187Networks"]
 
@@ -42,7 +46,7 @@ def load_atlas_setting(atlas: Union[str, Path, dict]):
         Path to the atlas files.
     """
     atlas_config = _check_altas_config(atlas)
-    print(atlas_config)
+    gc_log.info(atlas_config)
 
     # load template flow
     templateflow_dir = atlas_config.get("templateflow_dir")
@@ -104,7 +108,7 @@ def resample_atlas_collection(
     List of pathlib.Path
         Paths to atlases sampled to group level grey matter mask.
     """
-    print("Resample atlas to group grey matter mask.")
+    gc_log.info("Resample atlas to group grey matter mask.")
     resampled_atlases = []
     for desc in tqdm(atlas_config["file_paths"]):
         parcellation = atlas_config["file_paths"][desc]

giga_connectome/logger.py

@@ -0,0 +1,16 @@
+"""General logger for the cohort_creator package."""
+from __future__ import annotations
+
+import logging
+
+
+def gc_logger(log_level: str = "INFO") -> logging.Logger:
+    FORMAT = "%(message)s"
+
+    logging.basicConfig(
+        level=log_level,
+        format=FORMAT,
+        datefmt="[%X]",
+    )
+
+    return logging.getLogger("giga_connectome")

giga_connectome/mask.py

@@ -19,6 +19,10 @@
 
 from giga_connectome.atlas import resample_atlas_collection
 
+from giga_connectome.logger import gc_logger
+
+gc_log = gc_logger()
+
 
 def generate_gm_mask_atlas(
     working_dir: Path,
@@ -63,7 +67,6 @@ def generate_group_mask(
     template: str = "MNI152NLin2009cAsym",
     templateflow_dir: Optional[Path] = None,
     n_iter: int = 2,
-    verbose: int = 1,
 ) -> Nifti1Image:
     """
     Generate a group EPI grey matter mask, and overlaid with a MNI grey
@@ -88,9 +91,6 @@ def generate_group_mask(
         Number of repetitions of dilation and erosion steps performed in
         scipy.ndimage.binary_closing function.
 
-    verbose :
-        Level of verbosity.
-
     Keyword Arguments
     -----------------
     Used to filter the cirret
@@ -102,12 +102,10 @@ def generate_group_mask(
     nibabel.nifti1.Nifti1Image
         EPI (grey matter) mask for the current group of subjects.
     """
-    if verbose > 1:
-        print(f"Found {len(imgs)} masks")
-    if exclude := _check_mask_affine(imgs, verbose):
+    gc_log.debug(f"Found {len(imgs)} masks")
+    if exclude := _check_mask_affine(imgs):
         imgs, __annotations__ = _get_consistent_masks(imgs, exclude)
-        if verbose > 1:
-            print(f"Remaining: {len(imgs)} masks")
+        gc_log.debug(f"Remaining: {len(imgs)} masks")
 
     # templateflow environment setting to get around network issue
     if templateflow_dir and templateflow_dir.exists():
@@ -129,7 +127,7 @@ def generate_group_mask(
         memory=None,
         verbose=0,
     )
-    print(
+    gc_log.info(
         f"Group EPI mask affine:\n{group_epi_mask.affine}"
         f"\nshape: {group_epi_mask.shape}"
     )
@@ -204,7 +202,7 @@ def _get_consistent_masks(
 
 
 def _check_mask_affine(
-    mask_imgs: List[Union[Path, str, Nifti1Image]], verbose: int = 1
+    mask_imgs: List[Union[Path, str, Nifti1Image]]
 ) -> Union[list, None]:
     """Given a list of input mask images, show the most common affine matrix
     and subjects with different values.
@@ -215,9 +213,6 @@ def _check_mask_affine(
         See :ref:`extracting_data`.
         3D or 4D EPI image with same affine.
 
-    verbose :
-        Level of verbosity.
-
     Returns
     -------
 
@@ -244,12 +239,11 @@ def _check_mask_affine(
     common_affine = max(
         set(header_info["affine"]), key=header_info["affine"].count
     )
-    if verbose > 0:
-        print(
-            f"We found {len(set(header_info['affine']))} unique affine "
-            f"matrices. The most common one is "
-            f"{key_to_header[common_affine]}"
-        )
+    gc_log.debug(
+        f"We found {len(set(header_info['affine']))} unique affine "
+        f"matrices. The most common one is "
+        f"{key_to_header[common_affine]}"
+    )
     odd_balls = set(header_info["affine"]) - {common_affine}
     if not odd_balls:
         return None
@@ -259,18 +253,16 @@ def _check_mask_affine(
         ob_index = [
             i for i, aff in enumerate(header_info["affine"]) if aff == ob
         ]
-        if verbose > 1:
-            print(
-                "The following subjects has a different affine matrix "
-                f"({key_to_header[ob]}) comparing to the most common value: "
-                f"{mask_imgs[ob_index]}."
-            )
-        exclude += ob_index
-    if verbose > 0:
-        print(
-            f"{len(exclude)} out of {len(mask_imgs)} has "
-            "different affine matrix. Ignore when creating group mask."
+        gc_log.debug(
+            "The following subjects has a different affine matrix "
+            f"({key_to_header[ob]}) comparing to the most common value: "
+            f"{mask_imgs[ob_index]}."
         )
+        exclude += ob_index
+    gc_log.info(
+        f"{len(exclude)} out of {len(mask_imgs)} has "
+        "different affine matrix. Ignore when creating group mask."
+    )
     return sorted(exclude)
 
 
@@ -284,7 +276,9 @@ def _check_pregenerated_masks(template, working_dir, atlas):
     if not group_mask.exists():
         group_mask = None
     else:
-        print(f"Found pregenerated group level grey matter mask: {group_mask}")
+        gc_log.info(
+            f"Found pregenerated group level grey matter mask: {group_mask}"
+        )
 
     # atlas
     resampled_atlases = []
@@ -301,7 +295,7 @@ def _check_pregenerated_masks(template, working_dir, atlas):
     if not all(all_exist):
         resampled_atlases = None
     else:
-        print(
+        gc_log.info(
             f"Found resampled atlases: {resampled_atlases}. Skipping group "
             "level mask generation step."
         )

giga_connectome/postprocess.py

@@ -7,9 +7,13 @@
 from nilearn.connectome import ConnectivityMeasure
 from nilearn.maskers import NiftiLabelsMasker, NiftiMapsMasker
 from bids.layout import BIDSImageFile
+
 from giga_connectome import utils
 from giga_connectome.connectome import generate_timeseries_connectomes
 from giga_connectome.denoise import denoise_nifti_voxel
+from giga_connectome.logger import gc_logger
+
+gc_log = gc_logger()
 
 
 def run_postprocessing_dataset(
@@ -97,7 +101,8 @@ def run_postprocessing_dataset(
     )
 
     # transform data
-    print("processing subjects")
+    gc_log.info("processing subjects")
+
     for img in tqdm(images):
         # process timeseries
         denoised_img = denoise_nifti_voxel(
@@ -109,7 +114,9 @@ def run_postprocessing_dataset(
             attribute_name = f"{subject}_{specifier}_atlas-{atlas}_desc-{desc}"
             if not denoised_img:
                 time_series_atlas, correlation_matrix = None, None
-                print(f"{attribute_name}: no volume after scrubbing")
+
+                gc_log.info(f"{attribute_name}: no volume after scrubbing")
+
                 continue
 
             # extract timeseries and connectomes
@@ -140,7 +147,9 @@ def run_postprocessing_dataset(
                 )
 
     if analysis_level == "group":
-        print("create group connectome")
+
+        gc_log.info("create group connectome")
+
         for desc in connectomes:
             average_connectome = np.mean(
                 np.array(connectomes[desc]), axis=0

giga_connectome/run.py

@@ -105,6 +105,17 @@ def main(argv=None):
         "pipeline (option A). The default is False.",
         action="store_true",
     )
+    parser.add_argument(
+        "--verbosity",
+        help="""
+        Verbosity level.
+        """,
+        required=False,
+        choices=[0, 1, 2, 3],
+        default=2,
+        type=int,
+        nargs=1,
+    )
 
     args = parser.parse_args(argv)
 

giga_connectome/tests/test_mask.py

@@ -33,8 +33,7 @@ def test_check_mask_affine():
     weird = Nifti1Image(processed_vol, np.eye(4) * np.array([1, 1, 1.5, 1]).T)
     weird2 = Nifti1Image(processed_vol, np.eye(4) * np.array([1, 1, 1.6, 1]).T)
     exclude = mask._check_mask_affine(
-        [processed, processed, processed, processed, weird, weird, weird2],
-        verbose=2,
+        [processed, processed, processed, processed, weird, weird, weird2]
     )
     assert len(exclude) == 3
     assert exclude == [4, 5, 6]

giga_connectome/utils.py

@@ -4,6 +4,10 @@
 from bids.layout import Query
 from bids import BIDSLayout
 
+from giga_connectome.logger import gc_logger
+
+gc_log = gc_logger()
+
 
 def get_bids_images(
     subjects: List[str],
@@ -175,7 +179,7 @@ def get_subject_lists(
     ]
 
 
-def check_path(path: Path, verbose=True):
+def check_path(path: Path):
     """Check if given path (file or dir) already exists, and if so returns a
     new path with _<n> appended (n being the number of paths with the same name
     that exist already).
@@ -196,6 +200,7 @@ def check_path(path: Path, verbose=True):
         ]
         n = str(max(existing_numbers) + 1) if existing_numbers else "1"
         path = path_parent / f"{path.stem}_{n}{ext}"
-        if verbose:
-            print(f"Specified path already exists, using {path} instead.")
+
+        gc_log.debug(f"Specified path already exists, using {path} instead.")
+
     return path

giga_connectome/workflow.py

@@ -10,10 +10,29 @@
 
 from giga_connectome.denoise import is_ica_aroma
 from giga_connectome import utils
+from giga_connectome.logger import gc_logger
+
+
+gc_log = gc_logger()
+
+
+def set_verbosity(verbosity: int | list[int]) -> None:
+    if isinstance(verbosity, list):
+        verbosity = verbosity[0]
+    if verbosity == 0:
+        gc_log.setLevel("ERROR")
+    elif verbosity == 1:
+        gc_log.setLevel("WARNING")
+    elif verbosity == 2:
+        gc_log.setLevel("INFO")
+    elif verbosity == 3:
+        gc_log.setLevel("DEBUG")
 
 
 def workflow(args):
-    print(vars(args))
+
+    gc_log.info(vars(args))
+
     # set file paths
     bids_dir = args.bids_dir
     output_dir = args.output_dir
@@ -30,6 +49,8 @@ def workflow(args):
     strategy = get_denoise_strategy(args.denoise_strategy)
     atlas = load_atlas_setting(args.atlas)
 
+    set_verbosity(args.verbosity)
+
     # check output path
     output_dir.mkdir(parents=True, exist_ok=True)
     working_dir.mkdir(parents=True, exist_ok=True)
@@ -38,7 +59,8 @@ def workflow(args):
     template = (
         "MNI152NLin6Asym" if is_ica_aroma(strategy) else "MNI152NLin2009cAsym"
     )
-    print("Indexing BIDS directory")
+
+    gc_log.info("Indexing BIDS directory")
 
     # create subject ts and connectomes
     # refactor the two cases into one
@@ -55,8 +77,10 @@ def workflow(args):
                 f"sub-{subject}_atlas-{atlas['name']}"
                 f"_desc-{strategy['name']}.h5"
             )
-            connectome_path = utils.check_path(connectome_path, verbose=True)
-            print("Generate subject level connectomes")
+            connectome_path = utils.check_path(connectome_path)
+
+            gc_log.info("Generate subject level connectomes")
+
             run_postprocessing_dataset(
                 strategy,
                 resampled_atlases,
@@ -80,9 +104,11 @@ def workflow(args):
     connectome_path = (
         output_dir / f"atlas-{atlas['name']}_desc-{strategy['name']}.h5"
     )
-    connectome_path = utils.check_path(connectome_path, verbose=True)
-    print(connectome_path)
-    print("Generate subject level connectomes")
+    connectome_path = utils.check_path(connectome_path)
+
+    gc_log.info(connectome_path)
+    gc_log.info("Generate subject level connectomes")
+
     run_postprocessing_dataset(
         strategy,
         resampled_atlases,

tools/download_templates.py

@@ -4,6 +4,10 @@
 """
 from pathlib import Path
 
+from giga_connectome.logger import gc_logger
+
+gc_log = gc_logger()
+
 
 def fetch_tpl_atlas():
     """Download datasets from templateflow."""
@@ -13,7 +17,7 @@ def fetch_tpl_atlas():
     for atlas in atlases:
         tf_path = tf.get("MNI152NLin2009cAsym", atlas=atlas)
         if isinstance(tf_path, list) and len(tf_path) > 0:
-            print(f"{atlas} exists.")
+            gc_log.info(f"{atlas} exists.")
     # download MNI grey matter template
     tf.get("MNI152NLin2009cAsym", label="GM")
 
@@ -24,7 +28,7 @@ def download_mist():
 
     tf_path = templateflow.api.get("MNI152NLin2009bAsym", atlas="BASC")
     if isinstance(tf_path, list) and len(tf_path) > 0:
-        print("BASC / MIST atlas exists.")
+        gc_log.info("BASC / MIST atlas exists.")
         return
 
     # download and convert