Added Tutorial #9

cache.py

@@ -0,0 +1,156 @@
+########################################################################
+#
+# Cache-wrapper for a function or class.
+#
+# Save the result of calling a function or creating an object-instance
+# to harddisk. This is used to persist the data so it can be reloaded
+# very quickly and easily.
+#
+# Implemented in Python 3.5
+#
+########################################################################
+#
+# This file is part of the TensorFlow Tutorials available at:
+#
+# https://github.com/Hvass-Labs/TensorFlow-Tutorials
+#
+# Published under the MIT License. See the file LICENSE for details.
+#
+# Copyright 2016 by Magnus Erik Hvass Pedersen
+#
+########################################################################
+
+import os
+import pickle
+import numpy as np
+
+########################################################################
+
+
+def cache(cache_path, fn, *args, **kwargs):
+    """
+    Cache-wrapper for a function or class. If the cache-file exists
+    then the data is reloaded and returned, otherwise the function
+    is called and the result is saved to cache. The fn-argument can
+    also be a class instead, in which case an object-instance is
+    created and saved to the cache-file.
+
+    :param cache_path:
+        File-path for the cache-file.
+
+    :param fn:
+        Function or class to be called.
+
+    :param args:
+        Arguments to the function or class-init.
+
+    :param kwargs:
+        Keyword arguments to the function or class-init.
+
+    :return:
+        The result of calling the function or creating the object-instance.
+    """
+
+    # If the cache-file exists.
+    if os.path.exists(cache_path):
+        # Load the cached data from the file.
+        with open(cache_path, mode='rb') as file:
+            obj = pickle.load(file)
+
+        print("- Data loaded from cache-file: " + cache_path)
+    else:
+        # The cache-file does not exist.
+
+        # Call the function / class-init with the supplied arguments.
+        obj = fn(*args, **kwargs)
+
+        # Save the data to a cache-file.
+        with open(cache_path, mode='wb') as file:
+            pickle.dump(obj, file)
+
+        print("- Data saved to cache-file: " + cache_path)
+
+    return obj
+
+
+########################################################################
+
+
+def convert_numpy2pickle(in_path, out_path):
+    """
+    Convert a numpy-file to pickle-file.
+
+    The first version of the cache-function used numpy for saving the data.
+    Instead of re-calculating all the data, you can just convert the
+    cache-file using this function.
+
+    :param in_path:
+        Input file in numpy-format written using numpy.save().
+
+    :param out_path:
+        Output file written as a pickle-file.
+
+    :return:
+        Nothing.
+    """
+
+    # Load the data using numpy.
+    data = np.load(in_path)
+
+    # Save the data using pickle.
+    with open(out_path, mode='wb') as file:
+        pickle.dump(data, file)
+
+
+########################################################################
+
+if __name__ == '__main__':
+    # This is a short example of using a cache-file.
+
+    # This is the function that will only get called if the result
+    # is not already saved in the cache-file. This would normally
+    # be a function that takes a long time to compute, or if you
+    # need persistent data for some other reason.
+    def expensive_function(a, b):
+        return a * b
+
+    print('Computing expensive_function() ...')
+
+    # Either load the result from a cache-file if it already exists,
+    # otherwise calculate expensive_function(a=123, b=456) and
+    # save the result to the cache-file for next time.
+    result = cache(cache_path='cache_expensive_function.pkl',
+                   fn=expensive_function, a=123, b=456)
+
+    print('result =', result)
+
+    # Newline.
+    print()
+
+    # This is another example which saves an object to a cache-file.
+
+    # We want to cache an object-instance of this class.
+    # The motivation is to do an expensive computation only once,
+    # or if we need to persist the data for some other reason.
+    class ExpensiveClass:
+        def __init__(self, c, d):
+            self.c = c
+            self.d = d
+            self.result = c * d
+
+        def print_result(self):
+            print('c =', self.c)
+            print('d =', self.d)
+            print('result = c * d =', self.result)
+
+    print('Creating object from ExpensiveClass() ...')
+
+    # Either load the object from a cache-file if it already exists,
+    # otherwise make an object-instance ExpensiveClass(c=123, d=456)
+    # and save the object to the cache-file for the next time.
+    obj = cache(cache_path='cache_ExpensiveClass.pkl',
+                fn=ExpensiveClass, c=123, d=456)
+
+    obj.print_result()
+
+########################################################################

convert.py

@@ -0,0 +1,197 @@
+#!/usr/bin/python
+
+########################################################################
+#
+# Function and script for converting videos to images.
+#
+# This can be run as a script in a Linux shell by typing:
+#
+#    python convert.py
+#
+# Or by running:
+#
+#    chmod +x convert.py
+#    ./convert.py
+#
+# Requires the program avconv to be installed.
+# Tested with avconv v. 9.18-6 on Linux Mint.
+#
+# Implemented in Python 3.5 (seems to work in Python 2.7 as well)
+#
+########################################################################
+#
+# This file is part of the TensorFlow Tutorials available at:
+#
+# https://github.com/Hvass-Labs/TensorFlow-Tutorials
+#
+# Published under the MIT License. See the file LICENSE for details.
+#
+# Copyright 2016 by Magnus Erik Hvass Pedersen
+#
+########################################################################
+
+import os
+import subprocess
+import argparse
+
+########################################################################
+
+
+def video2images(in_dir, out_dir, crop_size, out_size, framerate, video_exts):
+    """
+    Convert videos to images. The videos are located in the directory in_dir
+    and all its sub-directories which are processed recursively. The directory
+    structure is replicated to out_dir where the jpeg-images are saved.
+
+    :param in_dir:
+        Input directory for the videos e.g. "/home/magnus/video/"
+        All sub-directories are processed recursively.
+
+    :param out_dir:
+        Output directory for the images e.g. "/home/magnus/video-images/"
+
+    :param crop_size:
+        Integer. First the videos are cropped to this width and height.
+
+    :param out_size:
+        Integer. After cropping, the videos are resized to this width and height.
+
+    :param framerate:
+        Integer. Number of frames to grab per second.
+
+    :param video_exts:
+        Tuple of strings. Extensions for video-files e.g. ('.mts', '.mp4')
+        Not case-sensitive.
+
+    :return:
+        Nothing.
+    """
+
+    # Convert all video extensions to lower-case.
+    video_exts = tuple(ext.lower() for ext in video_exts)
+
+    # Number of videos processed.
+    video_count = 0
+
+    # Process all the sub-dirs recursively.
+    for current_dir, dir_names, file_names in os.walk(in_dir):
+        # The current dir relative to the input directory.
+        relative_path = os.path.relpath(current_dir, in_dir)
+
+        # Name of the new directory for the output images.
+        new_dir = os.path.join(out_dir, relative_path)
+
+        # If the output-directory does not exist, then create it.
+        if not os.path.exists(new_dir):
+            os.makedirs(new_dir)
+
+        # For all the files in the current directory.
+        for file_name in file_names:
+            # If the file has a valid video-extension. Compare lower-cases.
+            if file_name.lower().endswith(video_exts):
+                # File-path for the input video.
+                in_file = os.path.join(current_dir, file_name)
+
+                # Split the file-path in root and extension.
+                file_root, file_ext = os.path.splitext(file_name)
+
+                # Create the template file-name for the output images.
+                new_file_name = file_root + "-%4d.jpg"
+
+                # Complete file-path for the output images incl. all sub-dirs.
+                new_file_path = os.path.join(new_dir, new_file_name)
+
+                # Clean up the path by removing e.g. "/./"
+                new_file_path = os.path.normpath(new_file_path)
+
+                # Print status.
+                print("Converting video to images:")
+                print("- Input video:   {0}".format(in_file))
+                print("- Output images: {0}".format(new_file_path))
+
+                # Command to be run in the shell for the video-conversion tool.
+                cmd = "avconv -i {0} -r {1} -vf crop={2}:{2} -vf scale={3}:{3} -qscale 2 {4}"
+
+                # Fill in the arguments for the command-line.
+                cmd = cmd.format(in_file, framerate, crop_size, out_size, new_file_path)
+
+                # Run the command-line in a shell.
+                subprocess.call(cmd, shell=True)
+
+                # Increase the number of videos processed.
+                video_count += 1
+
+                # Print newline.
+                print()
+
+    print("Number of videos converted: {0}".format(video_count))
+
+
+########################################################################
+# This script allows you to run the video-conversion from the command-line.
+
+if __name__ == "__main__":
+    # Argument description.
+    desc = "Convert videos to images. " \
+           "Recursively processes all sub-dirs of INDIR " \
+           "and replicates the dir-structure to OUTDIR. " \
+           "The video is first cropped to CROP:CROP pixels, " \
+           "then resized to SIZE:SIZE pixels and written as a jpeg-file. "
+
+    # Create the argument parser.
+    parser = argparse.ArgumentParser(description=desc)
+
+    # Add arguments to the parser.
+    parser.add_argument("--indir", required=True,
+                        help="input directory where videos are located")
+
+    parser.add_argument("--outdir", required=True,
+                        help="output directory where images will be saved")
+
+    parser.add_argument("--crop", required=True, type=int,
+                        help="the input videos are first cropped to CROP:CROP pixels")
+
+    parser.add_argument("--size", required=True, type=int,
+                        help="the input videos are then resized to SIZE:SIZE pixels")
+
+    parser.add_argument("--rate", required=False, type=int, default=5,
+                        help="the number of frames to convert per second")
+
+    parser.add_argument("--exts", required=False, nargs="+",
+                        help="list of extensions for video-files e.g. .mts .mp4")
+
+    # Parse the command-line arguments.
+    args = parser.parse_args()
+
+    # Get the arguments.
+    in_dir = args.indir
+    out_dir = args.outdir
+    crop_size = args.crop
+    out_size = args.size
+    framerate = args.rate
+    video_exts = args.exts
+
+    if video_exts is None:
+        # Default extensions for video-files.
+        video_exts = (".MTS", ".mp4")
+    else:
+        # A list of strings is provided as a command-line argument, but we
+        # need a tuple instead of a list, so convert it to a tuple.
+        video_exts = tuple(video_exts)
+
+    # Print the arguments.
+    print("Convert videos to images.")
+    print("- Input dir: " + in_dir)
+    print("- Output dir: " + out_dir)
+    print("- Crop width and height: {0}".format(crop_size))
+    print("- Resize width and height: {0}".format(out_size))
+    print("- Frame-rate: {0}".format(framerate))
+    print("- Video extensions: {0}".format(video_exts))
+    print()
+
+    # Perform the conversions.
+    video2images(in_dir=in_dir, out_dir=out_dir,
+                 crop_size=crop_size, out_size=out_size,
+                 framerate=framerate, video_exts=video_exts)
+
+########################################################################