Changes to --tedpca option

tedana/decomposition/pca.py

@@ -85,7 +85,7 @@ def tedpca(
         (stationary Gaussian) process and are ordered from most to least aggressive
         (see :footcite:p:`li2007estimating`).
         If a float is provided, then it is assumed to represent percentage of variance
-        explained (0-1) to retain from PCA.
+        explained (0.0-1.0) to retain from PCA.
         If an int is provided, then it is assumed to be the number of components
         to select
         Default is 'aic'.

tedana/tests/test_workflows_parser_utils.py

@@ -16,15 +16,29 @@ def test_check_tedpca_value():
         check_tedpca_value("hello", is_parser=False)
 
     with pytest.raises(argparse.ArgumentTypeError):
-        check_tedpca_value(1.5, is_parser=True)
+        check_tedpca_value("1.0", is_parser=True)
+
+    with pytest.raises(argparse.ArgumentTypeError):
+        check_tedpca_value("1.", is_parser=True)
+
+    with pytest.raises(ValueError):
+        check_tedpca_value(1.0, is_parser=False)
+
+    with pytest.raises(argparse.ArgumentTypeError):
+        check_tedpca_value("0.0", is_parser=True)
 
     with pytest.raises(ValueError):
-        check_tedpca_value(1.5, is_parser=False)
+        check_tedpca_value(0.0, is_parser=False)
 
     with pytest.raises(ValueError):
         check_tedpca_value(-1, is_parser=False)
 
-    assert check_tedpca_value(0.95) == 0.95
+    with pytest.raises(ValueError):
+        check_tedpca_value(0, is_parser=False)
+
+    assert check_tedpca_value(0.95, is_parser=False) == 0.95
     assert check_tedpca_value("0.95") == 0.95
     assert check_tedpca_value("mdl") == "mdl"
-    assert check_tedpca_value(52) == 52
+    assert check_tedpca_value(1, is_parser=False) == 1
+    assert check_tedpca_value("1") == 1
+    assert check_tedpca_value(52, is_parser=False) == 52

tedana/workflows/parser_utils.py

@@ -4,33 +4,54 @@
 import os.path as op
 
 
-def check_tedpca_value(string, is_parser=True):
+def check_tedpca_value(value, is_parser=True):
     """
     Check tedpca argument.
 
-    Check if argument is a float in range (0,1), an int greater than 1 or one of a
-    list of strings.
+    Check if argument is a float in range (0,1),
+    a positive integer,
+    or one of a list of strings.
     """
     valid_options = ("mdl", "aic", "kic", "kundu", "kundu-stabilize")
-    if string in valid_options:
-        return string
+    if value in valid_options:
+        return value
 
     error = argparse.ArgumentTypeError if is_parser else ValueError
+    dashes = "--" if is_parser else ""
+
+    def check_float(floatvalue):
+        assert isinstance(floatvalue, float)
+        if not (0.0 < floatvalue < 1.0):
+            msg = f"Floating-point argument to {dashes}tedpca must be in the range (0.0, 1.0)."
+            raise error(msg)
+        return floatvalue
+
+    def check_int(intvalue):
+        assert isinstance(intvalue, int)
+        if intvalue < 1:
+            msg = f"Integer argument to {dashes}tedpca must be positive"
+            raise error(msg)
+        return intvalue
+
+    if isinstance(value, float):
+        return check_float(value)
+    if isinstance(value, int):
+        return check_int(value)
+
     try:
-        floatarg = float(string)
+        floatvalue = float(value)
     except ValueError:
-        msg = f"Argument to tedpca must be a number or one of: {', '.join(valid_options)}"
+        msg = (
+            f"Argument to {dashes}tedpca must be either a number, "
+            f"or one of: {', '.join(valid_options)}"
+        )
         raise error(msg)
 
-    if floatarg != int(floatarg):
-        if not (0 < floatarg < 1):
-            raise error("Float argument to tedpca must be between 0 and 1.")
-        return floatarg
-    else:
-        intarg = int(floatarg)
-        if floatarg < 1:
-            raise error("Int argument must be greater than 1")
-        return intarg
+    try:
+        intvalue = int(value)
+    except ValueError:
+        return check_float(floatvalue)
+    return check_int(intvalue)
 
 
 def is_valid_file(parser, arg):

tedana/workflows/tedana.py

@@ -148,17 +148,20 @@ def _get_parser():
         dest="tedpca",
         type=check_tedpca_value,
         help=(
-            "Method with which to select components in TEDPCA. "
+            "Method by which to select number of components in TEDPCA. "
+            "This can be one of the following: "
+            "String ('mdl', 'kic', 'aic', 'kundu', or 'kundu-stabilize'); "
+            "floating-point value in the range (0.0, 1.0); "
+            "positive integer value. "
             "PCA decomposition with the mdl, kic and aic options "
-            "is based on a Moving Average (stationary Gaussian) "
-            "process and are ordered from most to least aggressive. "
-            "'kundu' or 'kundu-stabilize' are selection methods that "
-            "were distributed with MEICA. "
-            "Users may also provide a float from 0 to 1, "
-            "in which case components will be selected based on the "
-            "cumulative variance explained or an integer greater than 1"
-            "in which case the specificed number of components will be "
-            "selected."
+            "are based on a Moving Average (stationary Gaussian) process, "
+            "and are ordered from most to least aggressive. "
+            "'kundu' or 'kundu-stabilize' are legacy selection methods "
+            "that were distributed with MEICA. "
+            "Floating-point inputs select components based on the "
+            "cumulative variance explained. "
+            "Integer inputs select the specificed number of components. "
+            "Default: 'aic'."
         ),
         default="aic",
     )
@@ -402,8 +405,8 @@ def tedana_workflow(
     tedpca : {'mdl', 'aic', 'kic', 'kundu', 'kundu-stabilize', float, int}, optional
         Method with which to select components in TEDPCA.
         If a float is provided, then it is assumed to represent percentage of variance
-        explained (0-1) to retain from PCA. If an int is provided, it will output
-        a fixed number of components defined by the integer between 1 and the
+        explained (0.0-1.0) to retain from PCA. If an int is provided, it will output
+        a fixed number of components defined by the integer; must be between 1 and the
         number of time points.
         Default is 'aic'.
     fixed_seed : :obj:`int`, optional
@@ -509,7 +512,7 @@ def tedana_workflow(
         gscontrol = [gscontrol]
 
     # Check value of tedpca *if* it is a predefined string,
-    # a float on [0, 1] or an int >= 1
+    # a float in (0.0, 1.0) or an int >= 1
     tedpca = check_tedpca_value(tedpca, is_parser=False)
 
     # For z-catted files, make sure it's a list of size 1