Initial commit

Author: matthew-brett

.gitignore

@@ -0,0 +1,2 @@
+*.pyc
+__pycache__/

README.md

@@ -0,0 +1,39 @@
+# Diagnostics project
+
+Script go in the `scripts` directory.
+
+Library code (such as Python modules or packages) goes in the `packages` directory.
+
+You should put this `packages` directory on your Python PATH.
+
+This 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 ..
+
+## Check the data
+
+    python3 scripts/validate_data.py data
+
+## Find outliers
+
+    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>, ...
+
+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

data/.gitignore

@@ -0,0 +1,4 @@
+# Tell git to ignore everything
+*
+# Except the data hash list
+!data_hashes.txt

data/data_hashes.txt

@@ -0,0 +1,20 @@
+fd0654cf3a93e9e4dc6ab535cdc9239d4e0618ec group00_sub01_run1.nii
+d0f0590489efbc0e489f3098394cd3d5d6ca158e group00_sub01_run2.nii
+46162b0151127fd81934a5a13db5a8f272b3d5d7 group00_sub02_run1.nii
+92586788b1476507c2ce04acda37a1532e5b514d group00_sub02_run2.nii
+28895f3a338e941b5c206ddcb458094e009b3ee6 group00_sub03_run1.nii
+7b58b334e81d186595c0fc99d1b836ab0ca83aa4 group00_sub03_run2.nii
+79e5f243b1b0fd35d7f1635239459bcebc6bb21c group00_sub04_run1.nii
+b9b154f8cfa8f66e1ebd8ae9883e0b75793eea5e group00_sub04_run2.nii
+354950c19598e1c139cf2006e00d0a548e4108b3 group00_sub05_run1.nii
+672221403ed6b5986fbefdd937c4fb1c189b26c6 group00_sub05_run2.nii
+b265ffe2fc753f3afa55aab50cde75230a9378bc group00_sub06_run1.nii
+bc720346588b5cee301604a3fbd875067d6bdfe2 group00_sub06_run2.nii
+5183bbdc8f7705e948c4641a06659003dd4036d9 group00_sub07_run1.nii
+ac965ec6b7c76f95f0b43d4683fa23d9e34cf702 group00_sub07_run2.nii
+522d66c31244943137e8ef07607f404bb164d571 group00_sub08_run1.nii
+8e801bc75b7191d5c2e840f050e153d4e7333e40 group00_sub08_run2.nii
+69cec23b3066460f9e2d7ed54a0655248ac3256d group00_sub09_run1.nii
+78d71f77ed58621e24ebb648a06468d8849f3e0c group00_sub09_run2.nii
+ba6da5b248071651a687216a8f3fcfbcada3df11 group00_sub10_run1.nii
+8cd9670e6f2b2a70b6faf4c5aa4cb82c9bee9a84 group00_sub10_run2.nii

scripts/find_outliers.py

@@ -0,0 +1,42 @@
+""" Python script to find outliers
+
+Run as:
+
+    python3 scripts/find_outliers.py data
+"""
+
+import sys
+
+def find_outliers(data_directory):
+    """ Print filenames and outlier indices for images in `data_directory`.
+
+    Print filenames and detected outlier indices to the terminal.
+
+    Parameters
+    ----------
+    data_directory : str
+        Directory containing containing images.
+
+    Returns
+    -------
+    None
+    """
+    # Your code here
+    raise RuntimeError('No code yet')
+
+
+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)
+
+
+if __name__ == '__main__':
+    # Python is running this file as a script, not importing it.
+    main()

scripts/validate_data.py

@@ -0,0 +1,70 @@
+""" Python script to validate data
+
+Run as:
+
+    python3 scripts/validata_data.py data
+"""
+
+import os
+import sys
+import hashlib
+
+def file_hash(filename):
+    """ Get byte contents of file `filename`, return SHA1 hash
+
+    Parameters
+    ----------
+    filename : str
+        Name of file to read
+
+    Returns
+    -------
+    hash : str
+        SHA1 hexadecimal hash string for contents of `filename`.
+    """
+    # Open the file, read contents as bytes.
+    # Calculate, return SHA1 has on the bytes from the file.
+    raise RuntimeError('No code yet')
+
+
+def validate_data(data_directory):
+    """ Read ``data_hashes.txt`` file in `data_directory`, check hashes
+
+    Parameters
+    ----------
+    data_directory : str
+        Directory containing data and ``data_hashes.txt`` file.
+
+    Returns
+    -------
+    None
+
+    Raises
+    ------
+    ValueError:
+        If hash value for any file is different from hash value recorded in
+        ``data_hashes.txt`` file.
+    """
+    # Read lines from ``data_hashes.txt`` file.
+    # Split into SHA1 hash and filename
+    # Calculate actual hash for given filename.
+    # If hash for filename is not the same as the one in the file, raise
+    # ValueError
+    raise RuntimeError("No code yet")
+
+
+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
+    validate_data(data_directory)
+
+
+if __name__ == '__main__':
+    # Python is running this file as a script, not importing it.
+    main()

solutions/validate_data.py

@@ -0,0 +1,76 @@
+""" Python script to validate data
+
+Run as:
+
+    python3 scripts/validata_data.py data
+"""
+
+import os
+import sys
+import hashlib
+
+def file_hash(filename):
+    """ Get byte contents of file `filename`, return SHA1 hash
+
+    Parameters
+    ----------
+    filename : str
+        Name of file to read
+
+    Returns
+    -------
+    hash : str
+        SHA1 hexadecimal hash string for contents of `filename`.
+    """
+    # Open the file, read contents as bytes.
+    # Calculate, return SHA1 has on the bytes from the file.
+    with open(filename, 'rb') as fobj:
+        contents = fobj.read()
+    return hashlib.sha1(contents).hexdigest()
+
+
+def validate_data(data_directory):
+    """ Read ``data_hashes.txt`` file in `data_directory`, check hashes
+
+    Parameters
+    ----------
+    data_directory : str
+        Directory containing data and ``data_hashes.txt`` file.
+
+    Returns
+    -------
+    None
+
+    Raises
+    ------
+    ValueError:
+        If hash value for any file is different from hash value recorded in
+        ``data_hashes.txt`` file.
+    """
+    # Read lines from ``data_hashes.txt`` file.
+    for line in open(os.path.join(data_directory, 'data_hashes.txt'), 'rt'):
+        # Split into SHA1 hash and filename
+        hash, filename = line.strip().split()
+        # Calculate actual hash for given filename.
+        actual_hash = file_hash(os.path.join(data_directory, filename))
+        # If hash for filename is not the same as the one in the file, raise
+        # ValueError
+        if hash != actual_hash:
+            raise ValueError("Hash for {} does not match".format(filename))
+
+
+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
+    validate_data(data_directory)
+
+
+if __name__ == '__main__':
+    # Python is running this file as a script, not importing it.
+    main()