junifer 0.0.7.dev332__py3-none-any.whl → 0.0.8__py3-none-any.whl

junifer/configs/juseless/datagrabbers/ucla.py

@@ -4,15 +4,52 @@
 #          Leonard Sasse <l.sasse@fz-juelich.de>
 # License: AGPL
 
+from enum import StrEnum
 from pathlib import Path
-from typing import Union
+from typing import Annotated, Literal
+
+from pydantic import BeforeValidator
 
 from ....api.decorators import register_datagrabber
-from ....datagrabber import PatternDataGrabber
-from ....utils import raise_error
+from ....datagrabber import ConfoundsFormat, DataType, PatternDataGrabber
+from ....typing import DataGrabberPatterns
+from ....utils import ensure_list
+
+
+__all__ = ["JuselessUCLA", "UCLATask"]
+
+
+class UCLATask(StrEnum):
+    """Accepted UCLA tasks."""
 
+    REST = "rest"
+    BART = "bart"
+    BHT = "bht"
+    PAMENC = "pamenc"
+    PAMRET = "pamret"
+    SCAP = "scap"
+    TASKSWITCH = "taskswitch"
+    STOPSIGNAL = "stopsignal"
 
-__all__ = ["JuselessUCLA"]
+
+_types = Literal[
+    DataType.BOLD,
+    DataType.T1w,
+    DataType.VBM_CSF,
+    DataType.VBM_GM,
+    DataType.VBM_WM,
+]
+
+_tasks = Literal[
+    UCLATask.REST,
+    UCLATask.BART,
+    UCLATask.BHT,
+    UCLATask.PAMENC,
+    UCLATask.PAMRET,
+    UCLATask.SCAP,
+    UCLATask.TASKSWITCH,
+    UCLATask.STOPSIGNAL,
+]
 
 
 @register_datagrabber
@@ -23,118 +60,88 @@ class JuselessUCLA(PatternDataGrabber):
 
     Parameters
     ----------
-    datadir : str or Path, optional
+    datadir : Path, optional
         The directory where the dataset is stored.
         (default "/data/project/psychosis_thalamus/data/fmriprep").
     types: {"BOLD", "T1w", "VBM_CSF", "VBM_GM", "VBM_WM"} or \
            list of the options, optional
-        UCLA data types. If None, all available data types are selected.
-        (default None).
+        The data type(s) to grab.
     tasks : {"rest", "bart", "bht", "pamenc", "pamret", \
             "scap", "taskswitch", "stopsignal"} or \
-            list of the options or None, optional
-        UCLA task sessions. If None, all available task sessions are
-        selected (default None).
+            list of the options, optional
+        UCLA task sessions.
+        By default, all available task are selected.
 
     """
 
-    def __init__(
-        self,
-        datadir: Union[
-            str, Path
-        ] = "/data/project/psychosis_thalamus/data/fmriprep",
-        types: Union[str, list[str], None] = None,
-        tasks: Union[str, list[str], None] = None,
-    ) -> None:
-        # Declare all tasks
-        all_tasks = [
-            "rest",
-            "bart",
-            "bht",
-            "pamenc",
-            "pamret",
-            "scap",
-            "taskswitch",
-            "stopsignal",
-        ]
-        # Set default tasks
-        if tasks is None:
-            tasks = all_tasks
-        else:
-            # Convert single task into list
-            if isinstance(tasks, str):
-                tasks = [tasks]
-            # Verify valid tasks
-            for t in tasks:
-                if t not in all_tasks:
-                    raise_error(
-                        f"{t} is not a valid task in the UCLA dataset!"
-                    )
-        self.tasks = tasks
-        # The patterns
-        patterns = {
-            "BOLD": {
-                "pattern": (
-                    "{subject}/func/{subject}_task-{task}_bold_space-"
-                    "MNI152NLin2009cAsym_preproc.nii.gz"
-                ),
-                "space": "MNI152NLin2009cAsym",
-                "confounds": {
-                    "pattern": (
-                        "{subject}/func/{subject}_"
-                        "task-{task}_bold_confounds.tsv"
-                    ),
-                    "space": "fmriprep",
-                },
-            },
-            "T1w": {
-                "pattern": (
-                    "{subject}/anat/{subject}_"
-                    "T1w_space-MNI152NLin2009cAsym_preproc.nii.gz"
-                ),
-                "space": "MNI152NLin2009cAsym",
-            },
-            "VBM_CSF": {
-                "pattern": (
-                    "{subject}/anat/{subject}_T1w_space-"
-                    "MNI152NLin2009cAsym_class-CSF_probtissue.nii.gz"
-                ),
-                "space": "MNI152NLin2009cAsym",
-            },
-            "VBM_GM": {
-                "pattern": (
-                    "{subject}/anat/{subject}_T1w_space-"
-                    "MNI152NLin2009cAsym_class-GM_probtissue.nii.gz"
-                ),
-                "space": "MNI152NLin2009cAsym",
-            },
-            "VBM_WM": {
+    # the commented out uri leads to new open neuro dataset which does
+    # NOT have preprocessed data
+    # uri = "https://github.com/OpenNeuroDatasets/ds000030.git"
+    datadir: Path = Path("/data/project/psychosis_thalamus/data/fmriprep")
+    types: Annotated[_types | list[_types], BeforeValidator(ensure_list)] = [  # noqa: RUF012
+        DataType.BOLD,
+        DataType.T1w,
+        DataType.VBM_CSF,
+        DataType.VBM_GM,
+        DataType.VBM_WM,
+    ]
+    tasks: Annotated[
+        UCLATask | list[UCLATask], BeforeValidator(ensure_list)
+    ] = [  # noqa: RUF012
+        UCLATask.REST,
+        UCLATask.BART,
+        UCLATask.BHT,
+        UCLATask.PAMENC,
+        UCLATask.PAMRET,
+        UCLATask.SCAP,
+        UCLATask.TASKSWITCH,
+        UCLATask.STOPSIGNAL,
+    ]
+    patterns: DataGrabberPatterns = {  # noqa: RUF012
+        "BOLD": {
+            "pattern": (
+                "{subject}/func/{subject}_task-{task}_bold_space-"
+                "MNI152NLin2009cAsym_preproc.nii.gz"
+            ),
+            "space": "MNI152NLin2009cAsym",
+            "confounds": {
                 "pattern": (
-                    "{subject}/anat/{subject}_T1w_space"
-                    "-MNI152NLin2009cAsym_class-WM_probtissue.nii.gz"
+                    "{subject}/func/{subject}_task-{task}_bold_confounds.tsv"
                 ),
-                "space": "MNI152NLin2009cAsym",
+                "space": "fmriprep",
             },
-        }
-        # Set default types
-        if types is None:
-            types = list(patterns.keys())
-        # Convert single type into list
-        else:
-            if not isinstance(types, list):
-                types = [types]
-        # The replacements
-        replacements = ["subject", "task"]
-        # the commented out uri leads to new open neuro dataset which does
-        # NOT have preprocessed data
-        # uri = "https://github.com/OpenNeuroDatasets/ds000030.git"
-        super().__init__(
-            types=types,
-            datadir=datadir,
-            patterns=patterns,
-            replacements=replacements,
-            confounds_format="fmriprep",
-        )
+        },
+        "T1w": {
+            "pattern": (
+                "{subject}/anat/{subject}_"
+                "T1w_space-MNI152NLin2009cAsym_preproc.nii.gz"
+            ),
+            "space": "MNI152NLin2009cAsym",
+        },
+        "VBM_CSF": {
+            "pattern": (
+                "{subject}/anat/{subject}_T1w_space-"
+                "MNI152NLin2009cAsym_class-CSF_probtissue.nii.gz"
+            ),
+            "space": "MNI152NLin2009cAsym",
+        },
+        "VBM_GM": {
+            "pattern": (
+                "{subject}/anat/{subject}_T1w_space-"
+                "MNI152NLin2009cAsym_class-GM_probtissue.nii.gz"
+            ),
+            "space": "MNI152NLin2009cAsym",
+        },
+        "VBM_WM": {
+            "pattern": (
+                "{subject}/anat/{subject}_T1w_space"
+                "-MNI152NLin2009cAsym_class-WM_probtissue.nii.gz"
+            ),
+            "space": "MNI152NLin2009cAsym",
+        },
+    }
+    replacements: list[str] = ["subject", "task"]  # noqa: RUF012
+    confounds_format: ConfoundsFormat = ConfoundsFormat.FMRIPrep
 
     def get_elements(self) -> list:
         """Implement fetching list of elements in the dataset.

junifer/configs/juseless/datagrabbers/ukb_vbm.py

@@ -6,10 +6,13 @@
 # License: AGPL
 
 from pathlib import Path
-from typing import Union
+from typing import Literal
+
+from pydantic import AnyUrl
 
 from ....api.decorators import register_datagrabber
-from ....datagrabber import PatternDataladDataGrabber
+from ....datagrabber import DataType, PatternDataladDataGrabber
+from ....typing import DataGrabberPatterns
 
 
 __all__ = ["JuselessDataladUKBVBM"]
@@ -23,29 +26,20 @@ class JuselessDataladUKBVBM(PatternDataladDataGrabber):
 
     Parameters
     ----------
-    datadir : str or pathlib.Path or None, optional
-        The directory where the datalad dataset will be cloned. If None,
-        the datalad dataset will be cloned into a temporary directory
-        (default None).
+    datadir : pathlib.Path, optional
+        That path where the datalad dataset will be cloned.
+        If not specified, the datalad dataset will be cloned into a temporary
+        directory.
 
     """
 
-    def __init__(self, datadir: Union[str, Path, None] = None) -> None:
-        uri = "ria+http://ukb.ds.inm7.de#~cat_m0wp1"
-        rootdir = "m0wp1"
-        types = ["VBM_GM"]
-        replacements = ["subject", "session"]
-        patterns = {
-            "VBM_GM": {
-                "pattern": "m0wp1{subject}_ses-{session}_T1w.nii.gz",
-                "space": "IXI549Space",
-            },
-        }
-        super().__init__(
-            types=types,
-            datadir=datadir,
-            uri=uri,
-            rootdir=rootdir,
-            replacements=replacements,
-            patterns=patterns,
-        )
+    uri: AnyUrl = AnyUrl("ria+http://ukb.ds.inm7.de#~cat_m0wp1")
+    rootdir: Path = Path("m0wp1")
+    types: list[Literal[DataType.VBM_GM]] = [DataType.VBM_GM]  # noqa: RUF012
+    patterns: DataGrabberPatterns = {  # noqa: RUF012
+        "VBM_GM": {
+            "pattern": "m0wp1{subject}_ses-{session}_T1w.nii.gz",
+            "space": "IXI549Space",
+        },
+    }
+    replacements: list[str] = ["subject", "session"]  # noqa: RUF012

junifer/conftest.py

@@ -4,14 +4,64 @@
 #          Synchon Mandal <s.mandal@fz-juelich.de>
 # License: AGPL
 
+import gc
 from pathlib import Path
 
 import pytest
 
-from junifer.datagrabber import PatternDataladDataGrabber
 from junifer.utils.singleton import Singleton
 
 
+# Vendored packages, whose tests are not collected. Paths are relative to
+# this file, so it also works when testing the installed package.
+_VENDORED = [
+    Path(__file__).parent / "external" / "BrainPrint",
+]
+
+
+# TODO: replace with ``collect_ignore`` once datalad/datalad#7967 is fixed
+# (see juaml/junifer#518)
+def pytest_ignore_collect(collection_path: Path) -> bool | None:
+    """Do not collect the tests of vendored packages.
+
+    This replaces ``collect_ignore``, which is not applied when datalad's
+    pytest plugin (datalad >= 1.4) is installed: its ``pytest_ignore_collect``
+    returns False for directories, so pytest's own check is never called.
+
+    Parameters
+    ----------
+    collection_path : pathlib.Path
+        The path being considered for collection.
+
+    Returns
+    -------
+    bool or None
+        True to ignore the path, None to let other hooks decide.
+
+    """
+    if any(collection_path.is_relative_to(path) for path in _VENDORED):
+        return True
+    return None
+
+
+def pytest_collection_finish(session: pytest.Session) -> None:
+    """Exclude the objects alive after collection from garbage collection.
+
+    nilearn (< 0.14) forces a full garbage collection every time it gets
+    the data of an image, which has to go over all the objects of the test
+    session (modules, collected tests, etc.). Freezing them makes these
+    collections much faster.
+
+    Parameters
+    ----------
+    session : pytest.Session
+        The pytest session.
+
+    """
+    gc.collect()
+    gc.freeze()
+
+
 @pytest.fixture(autouse=True)
 def reset_singletons() -> None:
     """Reset all singletons."""
@@ -29,29 +79,18 @@ def reset_singletons() -> None:
         del elem
 
 
-@pytest.fixture
-def maps_datagrabber(tmp_path: Path) -> PatternDataladDataGrabber:
-    """Return a PatternDataladDataGrabber for maps testing.
+@pytest.fixture(autouse=True)
+def clear_computation_caches() -> None:
+    """Clear the cached computations of ReHo and ALFF estimators.
 
-    Parameters
-    ----------
-    tmp_path : pathlib.Path
-        The path to the test directory.
+    The estimators are singletons whose ``compute`` is cached per input
+    path, so results from earlier tests would otherwise be reused.
 
     """
-    dg = PatternDataladDataGrabber(
-        uri="https://github.com/OpenNeuroDatasets/ds005226.git",
-        types=["BOLD"],
-        patterns={
-            "BOLD": {
-                "pattern": (
-                    "derivatives/pre-processed_data/space-MNI/{subject}/"
-                    "{subject-padded}_task-{task}_run-{run}_space-MNI152NLin6Asym"
-                    "_res-2_desc-preproc_bold.nii.gz"
-                ),
-                "space": "MNI152NLin6Asym",
-            },
-        },
-        replacements=["subject", "subject-padded", "task", "run"],
-    )
-    return dg
+    from junifer.markers.falff._afni_falff import AFNIALFF
+    from junifer.markers.falff._junifer_falff import JuniferALFF
+    from junifer.markers.reho._afni_reho import AFNIReHo
+    from junifer.markers.reho._junifer_reho import JuniferReHo
+
+    for klass in (AFNIALFF, JuniferALFF, AFNIReHo, JuniferReHo):
+        klass.compute.cache_clear()

junifer/data/_dispatch.py

@@ -3,12 +3,11 @@
 # Authors: Synchon Mandal <s.mandal@fz-juelich.de>
 # License: AGPL
 
-from collections.abc import Iterator, MutableMapping
+from collections.abc import Callable, Iterator, MutableMapping
 from pathlib import Path
 from typing import (
     TYPE_CHECKING,
     Any,
-    Callable,
     Optional,
     Union,
 )
@@ -95,7 +94,7 @@ class DataDispatcher(MutableMapping):
         # Update global
         self._registries[key] = value
 
-    def popitem():
+    def popitem(self):
         """Not implemented."""
         pass
 
@@ -110,14 +109,9 @@ class DataDispatcher(MutableMapping):
 
 def get_data(
     kind: str,
-    names: Union[
-        str,  # coordinates, parcellation, mask
-        list[str],  # parcellation, mask
-        dict,  # mask
-        list[dict],  # mask
-    ],
+    names: str | list[str] | dict | list[dict],
     target_data: dict[str, Any],
-    extra_input: Optional[dict[str, Any]] = None,
+    extra_input: dict[str, Any] | None = None,
 ) -> Union[
     tuple[ArrayLike, list[str]],  # coordinates
     tuple["Nifti1Image", dict[int, str]],  # parcellation
@@ -196,15 +190,11 @@ def load_data(
     kind: str,
     name: str,
     **kwargs,
-) -> Union[
-    tuple[ArrayLike, list[str], str],  # coordinates
-    tuple[
-        Optional["Nifti1Image"], list[str], Path, str
-    ],  # parcellation / maps
-    tuple[
-        Optional[Union["Nifti1Image", Callable]], Optional[Path], str
-    ],  # mask
-]:
+) -> (
+    tuple[ArrayLike, list[str], str]
+    | tuple[Optional["Nifti1Image"], list[str], Path, str]
+    | tuple[Union["Nifti1Image", Callable] | None, Path | None, str]
+):
     """Load ``kind`` named ``name``.
 
     Parameters

junifer/data/coordinates/_ants_coordinates_warper.py

@@ -9,7 +9,7 @@ import numpy as np
 from numpy.typing import ArrayLike
 
 from ...pipeline import WorkDirManager
-from ...utils import logger, run_ext_cmd
+from ...utils import run_ext_cmd
 
 
 __all__ = ["ANTsCoordinatesWarper"]
@@ -46,6 +46,9 @@ class ANTsCoordinatesWarper:
             The transformed coordinates.
 
         """
+        # Imported here to avoid circular import
+        from ._coordinates import logger
+
         logger.debug("Using ANTs for coordinates transformation")
 
         # Create element-specific tempdir for storing post-warping assets

junifer/data/coordinates/_coordinates.py

@@ -4,15 +4,17 @@
 #          Synchon Mandal <s.mandal@fz-juelich.de>
 # License: AGPL
 
+from collections import Counter
 from pathlib import Path
-from typing import Any, Optional
+from typing import Any
 
 import numpy as np
 import pandas as pd
+import structlog
 from junifer_data import get
 from numpy.typing import ArrayLike
 
-from ...utils import logger, raise_error
+from ...utils import raise_error
 from ..pipeline_data_registry_base import BasePipelineDataRegistry
 from ..utils import JUNIFER_DATA_PARAMS, get_dataset_path, get_native_warper
 from ._ants_coordinates_warper import ANTsCoordinatesWarper
@@ -21,6 +23,49 @@ from ._fsl_coordinates_warper import FSLCoordinatesWarper
 
 __all__ = ["CoordinatesRegistry"]
 
+_log = structlog.get_logger("junifer")
+logger = _log.bind(pkg="data")
+
+
+def _sanitize_names(names: list) -> list:
+    """Make the VOI names safe to use as feature names.
+
+    Spaces in the names are replaced with ``_`` and duplicated names get a
+    ``-{i}`` suffix, where ``i`` is the (1-based) occurrence of the name,
+    skipping suffixes that would clash with another name. Non-string names
+    are left untouched.
+
+    Parameters
+    ----------
+    names : list
+        The names of the VOIs.
+
+    Returns
+    -------
+    list
+        The sanitized, unique names of the VOIs.
+
+    """
+    names = [x.replace(" ", "_") if isinstance(x, str) else x for x in names]
+    counts = Counter(names)
+    # Unique names are kept, so suffixed names must not clash with them
+    used = {name for name in names if counts[name] == 1}
+    next_idx: Counter = Counter()
+    out = []
+    for name in names:
+        if counts[name] > 1:
+            # Find the next suffix not in use
+            while True:
+                next_idx[name] += 1
+                new_name = f"{name}-{next_idx[name]}"
+                if new_name not in used:
+                    break
+            used.add(new_name)
+            out.append(new_name)
+        else:
+            out.append(name)
+    return out
+
 
 class CoordinatesRegistry(BasePipelineDataRegistry):
     """Class for coordinates data registry.
@@ -135,7 +180,7 @@ class CoordinatesRegistry(BasePipelineDataRegistry):
         coordinates: ArrayLike,
         voi_names: list[str],
         space: str,
-        overwrite: Optional[bool] = False,
+        overwrite: bool | None = False,
     ) -> None:
         """Register a custom user coordinates.
 
@@ -289,8 +334,8 @@ class CoordinatesRegistry(BasePipelineDataRegistry):
             )
             # Convert dataframe to numpy ndarray
             coords = df_coords.iloc[:, [0, 1, 2]].to_numpy()
-            # Get label names
-            names = list(df_coords.iloc[:, [3]].values[:, 0])
+            # Get label names, made unique and without spaces
+            names = _sanitize_names(list(df_coords.iloc[:, [3]].values[:, 0]))
         # Load data for external ones
         else:
             coords = t_coord["coords"]
@@ -302,7 +347,7 @@ class CoordinatesRegistry(BasePipelineDataRegistry):
         self,
         coords: str,
         target_data: dict[str, Any],
-        extra_input: Optional[dict[str, Any]] = None,
+        extra_input: dict[str, Any] | None = None,
     ) -> tuple[ArrayLike, list[str]]:
         """Get coordinates, tailored for the target data.
 

junifer/data/coordinates/_fsl_coordinates_warper.py

@@ -9,7 +9,7 @@ import numpy as np
 from numpy.typing import ArrayLike
 
 from ...pipeline import WorkDirManager
-from ...utils import logger, run_ext_cmd
+from ...utils import run_ext_cmd
 
 
 __all__ = ["FSLCoordinatesWarper"]
@@ -46,6 +46,9 @@ class FSLCoordinatesWarper:
             The transformed coordinates.
 
         """
+        # Imported here to avoid circular import
+        from ._coordinates import logger
+
         logger.debug("Using FSL for coordinates transformation")
 
         # Create element-specific tempdir for storing post-warping assets

junifer/data/coordinates/tests/test_coordinates.py

@@ -16,7 +16,10 @@ from junifer.data import (
     register_data,
 )
 from junifer.datareader import DefaultDataReader
-from junifer.testing.datagrabbers import OasisVBMTestingDataGrabber
+from junifer.testing.datagrabbers import UCLACNPVBMTestingDataGrabber
+
+
+pytestmark = pytest.mark.external
 
 
 def test_register_built_in_check() -> None:
@@ -121,6 +124,15 @@ def test_load() -> None:
     assert space == "MNI"
 
 
+def test_load_unique_names() -> None:
+    """Test loading coordinates with duplicated names and spaces."""
+    _, names, _ = load_data(kind="coordinates", name="Dosenbach")
+    assert len(names) == 160  # type: ignore
+    assert len(set(names)) == 160  # type: ignore
+    assert not any(" " in x for x in names)  # type: ignore
+    assert names[:5] == ["vmPFC-1", "aPFC-1", "aPFC-2", "mPFC", "aPFC-3"]
+
+
 def test_load_nonexisting() -> None:
     """Test loading coordinates that not exist."""
     with pytest.raises(ValueError, match=r"not found"):
@@ -130,8 +142,8 @@ def test_load_nonexisting() -> None:
 def test_get() -> None:
     """Test tailored coordinates fetch."""
     reader = DefaultDataReader()
-    with OasisVBMTestingDataGrabber() as dg:
-        element = dg["sub-01"]
+    with UCLACNPVBMTestingDataGrabber() as dg:
+        element = dg["sub-10206"]
         element_data = reader.fit_transform(element)
         vbm_gm = element_data["VBM_GM"]
         # Get tailored coordinates