Slightly more basic code

Author: matthew-brett

README.md

@@ -1,39 +1,49 @@
 # Diagnostics project
 
-Script go in the `scripts` directory.
+Scripts go in the `scripts` directory.
 
-Library code (such as Python modules or packages) goes in the `packages` directory.
+Library code (Python modules) goes in the `findoutlie` directory.
 
-You should put this `packages` directory on your Python PATH.
+You should put the code in this `findoutlie` directory on your Python PATH.
 
-This file has instructions on how to get, validate and process the data.
+This README file has instructions on how to get, validate and process the data.
 
 ## Get the data
 
-    cd data
-    curl -LO http://nipy.bic.berkeley.edu/psych-214/group00.tar.gz
-    tar zxvf group00.tar.gz
-    cd ..
+```
+cd data
+curl -L https://figshare.com/ndownloader/files/{DATA_ID} -o group_data.tar
+tar zxvf group.tar
+cd ..
+```
 
 ## Check the data
 
-    python3 scripts/validate_data.py data
+```
+python3 scripts/validate_data.py data
+```
 
 ## Find outliers
 
-    python3 scripts/find_outliers.py data
+```
+python3 scripts/find_outliers.py data
+```
 
 This should print output to the terminal of form:
 
-    <filename> <outlier_index>, <outlier_index>, ...
-    <filename> <outlier_index>, <outlier_index>, ...
+```
+<filename>, <outlier_index>, <outlier_index>, ...
+<filename>, <outlier_index>, <outlier_index>, ...
+```
 
 Where `<filename>` is the name of the image that has outlier scans, and
 `<outlier_index>` is an index to the volume in the 4D image that you have
-indentified as an outlier.  0 refers to the first volume.  For example:
-
-    group00_sub01_run1.nii 3, 21, 22, 104
-    group00_sub02_run2.nii 11, 33 91
-    group00_sub04_run2.nii 101, 102, 132
-    group00_sub07_run2.nii 0, 1, 2, 166, 167
-    group00_sub09_run2.nii 3
+identified as an outlier.  0 refers to the first volume.  For example:
+
+```
+data/sub-01/func/sub-01_task-taskzero_run-01_bold.nii.gz, 3, 21, 22, 104
+data/sub-01/func/sub-01_task-taskzero_run-02_bold.nii.gz, 11, 33, 91
+data/sub-03/func/sub-03_task-taskzero_run-02_bold.nii.gz, 101, 102, 132
+data/sub-08/func/sub-08_task-taskzero_run-01_bold.nii.gz, 0, 1, 2, 166, 167
+data/sub-09/func/sub-08_task-taskzero_run-01_bold.nii.gz, 3
+```

findoutlie/__init__.py

@@ -0,0 +1,32 @@
+""" Module with routines for finding outliers
+"""
+
+import os.path as op
+from glob import glob
+
+
+def detect_outliers(fname):
+    return [42]
+
+
+def find_outliers(data_directory):
+    """ Return filenames and outlier indices for images in `data_directory`.
+
+    Parameters
+    ----------
+    data_directory : str
+        Directory containing containing images.
+
+    Returns
+    -------
+    outlier_dict : dict
+        Dictionary with keys being filenames and values being lists of outliers
+        for filename.
+    """
+    image_fnames = glob(op.join(data_directory, '**', 'sub-*.nii.gz'),
+                        recursive=True)
+    outlier_dict = {}
+    for fname in image_fnames:
+        outliers = detect_outliers(fname)
+        outlier_dict[fname] = outliers
+    return outlier_dict

findoutlie/outfind.py

@@ -5,36 +5,45 @@
     python3 scripts/find_outliers.py data
 """
 
+import os.path as op
 import sys
 
-def find_outliers(data_directory):
-    """ Print filenames and outlier indices for images in `data_directory`.
+from argparse import ArgumentParser, RawDescriptionHelpFormatter
 
-    Print filenames and detected outlier indices to the terminal.
+# Put the findoutlie directory on the Python path.
+PACKAGE_DIR = op.join(op.dirname(__file__), '..')
+sys.path.append(PACKAGE_DIR)
 
-    Parameters
-    ----------
-    data_directory : str
-        Directory containing containing images.
+from findoutlie import outfind
 
-    Returns
-    -------
-    None
-    """
-    # Your code here
-    raise RuntimeError('No code yet')
+
+def print_outliers(data_directory):
+    outlier_dict = outfind.find_outliers(data_directory)
+    for fname, outliers in outlier_dict.items():
+        if len(outliers) == 0:
+            continue
+        outlier_strs = []
+        for out_ind in outliers:
+            outlier_strs.append(str(out_ind))
+        print(', '.join([fname] + outlier_strs))
+
+
+def get_parser():
+    parser = ArgumentParser(description=__doc__,  # Usage from docstring
+                            formatter_class=RawDescriptionHelpFormatter)
+    parser.add_argument('data_directory',
+                        help='Directory containing data')
+    return parser
 
 
 def main():
     # This function (main) called when this file run as a script.
     #
     # Get the data directory from the command line arguments
-    if len(sys.argv) < 2:
-        raise RuntimeError("Please give data directory on "
-                           "command line")
-    data_directory = sys.argv[1]
-    # Call function to validate data in data directory
-    find_outliers(data_directory)
+    parser = get_parser()
+    args = parser.parse_args()
+    # Call function to find outliers.
+    print_outliers(args.data_directory)
 
 
 if __name__ == '__main__':