From 0a3fd5b62065333669c7b391c626cb2505217617 Mon Sep 17 00:00:00 2001
From: Matt Cooper <matthew_cooper@brown.edu>
Date: Fri, 12 Aug 2016 16:48:46 -0400
Subject: First commit

---
 Code/process_data.py | 71 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 71 insertions(+)
 create mode 100644 Code/process_data.py

(limited to 'Code/process_data.py')

diff --git a/Code/process_data.py b/Code/process_data.py
new file mode 100644
index 0000000..170959a
--- /dev/null
+++ b/Code/process_data.py
@@ -0,0 +1,71 @@
+import numpy as np
+import getopt
+import sys
+from glob import glob
+
+import constants as c
+from utils import process_clip
+
+
+def process_training_data(num_clips):
+    """
+    Processes random training clips from the full training data. Saves to TRAIN_DIR_CLIPS by
+    default.
+
+    @param num_clips: The number of clips to process. Default = 5000000 (set in __main__).
+
+    @warning: This can take a couple of hours to complete with large numbers of clips.
+    """
+    num_prev_clips = len(glob(c.TRAIN_DIR_CLIPS + '*'))
+
+    for clip_num in xrange(num_prev_clips, num_clips + num_prev_clips):
+        clip = process_clip()
+
+        np.savez_compressed(c.TRAIN_DIR_CLIPS + str(clip_num), clip)
+
+        if (clip_num + 1) % 100 == 0: print 'Processed %d clips' % (clip_num + 1)
+
+
+def usage():
+    print 'Options:'
+    print '-n/--num_clips= <# clips to process for training>'
+    print '-t/--train_dir= <Directory of full training frames>'
+    print '-c/--clips_dir= <Save directory for processed clips>'
+    print "                (I suggest making this a hidden dir so the filesystem doesn't freeze"
+    print "                 with so many files. DON'T `ls` THIS DIRECTORY!)"
+    print '-o/--overwrite  (Overwrites the previous data in the training dir)'
+
+
+def main():
+    ##
+    # Handle command line input
+    ##
+
+    num_clips = 5000000
+
+    try:
+        opts, _ = getopt.getopt(sys.argv[1:], 'n:t:c:o',
+                                ['num_clips=', 'train_dir=', 'clips_dir=', 'overwrite'])
+    except getopt.GetoptError:
+        usage()
+        sys.exit(2)
+
+    for opt, arg in opts:
+        if opt in ('-n', '--num_clips'):
+            num_clips = int(arg)
+        if opt in ('-t', '--train_dir'):
+            c.TRAIN_DIR = c.get_dir(arg)
+        if opt in ('-c', '--clips_dir'):
+            c.TRAIN_DIR_CLIPS = c.get_dir(arg)
+        if opt in ('-o', '--overwrite'):
+            c.clear_dir(c.TRAIN_DIR_CLIPS)
+
+    ##
+    # Process data for training
+    ##
+
+    process_training_data(num_clips)
+
+
+if __name__ == '__main__':
+    main()
-- 
cgit v1.2.3-70-g09d2