loosen pandas version constraint from >=0.24.0 to >=0.21.0

>=0.21.0 required due to:
$ git diff v0.20.3 v0.21.0 pandas/core/generic.py
diff --git a/pandas/core/generic.py b/pandas/core/generic.py
index 87ce570e9..118e7d5cd 100644
--- a/pandas/core/generic.py
+++ b/pandas/core/generic.py
[...]
@@ -9,11 +9,10 @@ import json
 import numpy as np
 import pandas as pd
@@ -2126,14 +2445,23 @@ it is assumed to be aliases for the column names.')

         return self.reindex(**d)

-    def drop(self, labels, axis=0, level=None, inplace=False, errors='raise'):
+    def drop(self, labels=None, axis=0, index=None, columns=None, level=None,
+             inplace=False, errors='raise'):
         """
         Return new object with labels in requested axis removed.
[...]

http://pandas.pydata.org/pandas-docs/version/0.20/generated/pandas.DataFrame.drop.html
http://pandas.pydata.org/pandas-docs/version/0.21/generated/pandas.DataFrame.drop.html

freesurfer_volume_reader/__init__.py

@@ -55,9 +55,10 @@ def parse_hippocampal_volume_file_path(volume_file_path: str) -> dict:
 
 
 def read_hippocampal_volume_file_dataframe(volume_file_path: str) -> pandas.DataFrame:
-    volumes_frame = pandas.DataFrame(
-        read_hippocampal_volumes_mm3(volume_file_path).items(),
-        columns=['subfield', 'volume_mm^3'])
+    volumes_frame = pandas.DataFrame([
+        {'subfield': s, 'volume_mm^3': v}
+        for s, v in read_hippocampal_volumes_mm3(volume_file_path).items()
+    ])
     for key, value in parse_hippocampal_volume_file_path(volume_file_path).items():
         volumes_frame[key] = value
     # volumes_frame['hemisphere'] = volumes_frame['hemisphere'].astype('category')

setup.py

@@ -36,7 +36,8 @@ setuptools.setup(
         ],
     },
     install_requires=[
-        'pandas',
+        # pandas.DataFrame.drop(columns=[...], ...)
+        'pandas>=0.21.0',
     ],
     setup_requires=[
         'setuptools_scm',

tests/hippocampus_test.py

@@ -158,6 +158,7 @@ def assert_main_volume_frame_equals(capsys, argv: list, expected_frame: pandas.D
     out, _ = capsys.readouterr()
     assert_volume_frames_equal(
         left=expected_frame,
+        # pandas.DataFrame.drop(columns=[...], ...) >= pandas0.21.0
         right=pandas.read_csv(io.StringIO(out)).drop(columns=['source_path']),
     )