Merge branch 'main' into AddStaticTypeChecking

Author: oesteban

docs/notebooks/dmri_covariance.ipynb

@@ -68,6 +68,8 @@ resmon = ["psutil >=5.4"]
 
 test = [
     "coverage",
+    "ipykernel",
+    "mriqc-learn",
     "pytest >= 4.4",
     "pytest-cov",
     "pytest-env",
@@ -86,7 +88,7 @@ types = [
 antsopt = [
     "ConfigSpace",
     "nipreps",
-    "skimage",
+    "scikit-image",
     "smac",
 ]
 
@@ -201,6 +203,8 @@ filterwarnings = [
   "ignore:Updating b0_threshold to.*:UserWarning",
   # scikit-learn
   "ignore:The optimal value found for dimension.*:sklearn.exceptions.ConvergenceWarning",
+  # masks
+  "ignore:No mask provided;.*:UserWarning",
 ]
 
 

pyproject.toml

@@ -28,7 +28,7 @@
     TrivialModel,
 )
 from nifreeze.model.dmri import (
-    AverageDWModel,
+    AverageDWIModel,
     DKIModel,
     DTIModel,
     GPModel,
@@ -38,7 +38,7 @@
 __all__ = (
     "ModelFactory",
     "AverageModel",
-    "AverageDWModel",
+    "AverageDWIModel",
     "DKIModel",
     "DTIModel",
     "GPModel",

src/nifreeze/model/__init__.py

@@ -53,9 +53,9 @@ def init(model="DTI", **kwargs):
             return TrivialModel(predicted=kwargs.pop("S0"), gtab=kwargs.pop("gtab"))
 
         if model.lower() in ("avgdwi", "averagedwi", "meandwi"):
-            from nifreeze.model.dmri import AverageDWModel
+            from nifreeze.model.dmri import AverageDWIModel
 
-            return AverageDWModel(**kwargs)
+            return AverageDWIModel(**kwargs)
 
         if model.lower() in ("avg", "average", "mean"):
             return AverageModel(**kwargs)

src/nifreeze/model/base.py

@@ -21,6 +21,8 @@
 #     https://www.nipreps.org/community/licensing/
 #
 
+from importlib import import_module
+
 import numpy as np
 from joblib import Parallel, delayed
 
@@ -121,8 +123,6 @@ def __init__(self, gtab, S0=None, b_max=None, **kwargs):
         # DIPY models (or one with a fully-compliant interface)
         model_str = getattr(self, "_model_class", None)
         if model_str:
-            from importlib import import_module
-
             module_name, class_name = model_str.rsplit(".", 1)
             self._model = getattr(
                 import_module(module_name),
@@ -218,8 +218,8 @@ def predict(self, gradient=None, **kwargs):
         return retval
 
 
-class AverageDWModel(BaseDWIModel):
-    """A trivial model that returns an average map."""
+class AverageDWIModel(BaseDWIModel):
+    """A trivial model that returns an average DWI volume."""
 
     __slots__ = ("_data", "_th_low", "_th_high", "_bias", "_stat", "_is_fitted")
 

src/nifreeze/model/dmri.py

@@ -78,10 +78,10 @@ def test_average_model():
 
     data *= gtab[:, -1]
 
-    tmodel_mean = model.AverageDWModel(gtab=gtab, bias=False, stat="mean")
-    tmodel_median = model.AverageDWModel(gtab=gtab, bias=False, stat="median")
-    tmodel_1000 = model.AverageDWModel(gtab=gtab, bias=False, th_high=1000, th_low=900)
-    tmodel_2000 = model.AverageDWModel(
+    tmodel_mean = model.AverageDWIModel(gtab=gtab, bias=False, stat="mean")
+    tmodel_median = model.AverageDWIModel(gtab=gtab, bias=False, stat="median")
+    tmodel_1000 = model.AverageDWIModel(gtab=gtab, bias=False, th_high=1000, th_low=900)
+    tmodel_2000 = model.AverageDWIModel(
         gtab=gtab,
         bias=False,
         th_high=2000,
@@ -153,7 +153,7 @@ def test_two_initialisations(datadir):
     data_train, data_test = lovo_split(dmri_dataset, 10)
 
     # Direct initialisation
-    model1 = model.AverageDWModel(
+    model1 = model.AverageDWIModel(
         gtab=data_train[1],
         S0=dmri_dataset.bzero,
         th_low=100,