make compatible with python 3 (as well as 2)

model/lasagne_net.py

@@ -26,7 +26,7 @@
     from lasagne.layers import batch_norm as l_batch_norm 
 
 import config as cfg
-import lasagne_io as io
+from model import lasagne_io as io
 from utils import log
 
 from lasagne import random as lasagne_random

model/learning_rate.py

@@ -52,4 +52,4 @@ def dynamicLearningRate(mode, epoch):
 
     for epoch in range(1, cfg.EPOCHS + 1):
 
-        print dynamicLearningRate('cosine', epoch)
+        print(dynamicLearningRate('cosine', epoch))

sort_data.py

@@ -20,12 +20,12 @@ def parseDataset():
     # List of wav-files
     wav_path = os.path.join(cfg.TRAINSET_PATH, 'wav')
     wav_files = [f for f in sorted(os.listdir(wav_path))]
-    print 'DATASET CONTAINS', len(wav_files), 'WAV_FILES'
+    print('DATASET CONTAINS', len(wav_files), 'WAV_FILES')
 
     # List all xml-files
     xml_path = os.path.join(cfg.TRAINSET_PATH, 'xml')
     xml_files = [os.path.join(xml_path, f) for f in sorted(os.listdir(xml_path))]
-    print 'PARSING', len(xml_files), 'XML-FILES...'
+    print('PARSING', len(xml_files), 'XML-FILES...')
 
     # Open xml-files and extract metadata
     for i in range(len(xml_files)):
@@ -56,21 +56,21 @@ def parseDataset():
 
         # Status (parsing the files might take a while)
         if not i % 100:
-            print '\t', i, '/', len(xml_files)
+            print('\t', i, '/', len(xml_files))
 
-    print '...DONE!', len(metadata), 'CLASSES IN DATASET'
+    print('...DONE!', len(metadata), 'CLASSES IN DATASET')
 
     return metadata
 
 ####################  CREATE SPLITS  #####################
 def sortDataset(mdata):
 
-    print 'PARSING CLASSES...'
+    print('PARSING CLASSES...')
 
     # Parse classes
     for c in mdata:
 
-        print '\t', c
+        print('\t', c)
 
         # Determine size of val split (10% but at least 1 file)
         val = max(1, len(mdata[c]) * 0.1)
@@ -107,7 +107,7 @@ def sortDataset(mdata):
             else:
                 copyfile(os.path.join(cfg.TRAINSET_PATH, 'wav', f['filename']), os.path.join(t_path, f['filename']))
 
-    print '...DONE!'
+    print('...DONE!')
 
 
 if __name__ == '__main__':

spec.py

@@ -66,7 +66,7 @@ def parseDataset():
             try:
 
                 # Stats
-                print i + 1, '/', len(afiles), c, afiles[i],
+                print(i + 1, '/', len(afiles), c, afiles[i],)
 
                 # Get specs and signal to noise ratios
                 specs, noise = getSpecs(os.path.join(cfg.TRAINSET_PATH, 'train', c, afiles[i]))

submission_monophone.py

@@ -51,7 +51,7 @@ def getClassId(c):
     if c in LABELS:
         return CODES[LABELS.index(c)]
     else:
-        print 'MISSING CLASS:', c
+        print('MISSING CLASS:', c)
         return False       
 
 def runTest(SNAPSHOTS, TEST):

submission_soundscape.py

@@ -61,7 +61,7 @@ def getClassId(c):
     if c in LABELS:
         return CODES[LABELS.index(c)]
     else:
-        print 'MISSING CLASS:', c
+        print('MISSING CLASS:', c)
         return False
 
 def getSpecBatches(split):

utils/audio.py

@@ -2,6 +2,7 @@
 
 import numpy as np
 import librosa
+from builtins import range
 
 import cv2
 
@@ -20,7 +21,7 @@ def splitSignal(sig, rate, seconds, overlap, minlen):
 
     # Split signal with overlap
     sig_splits = []
-    for i in xrange(0, len(sig), int((seconds - overlap) * rate)):
+    for i in range(0, len(sig), int((seconds - overlap) * rate)):
         split = sig[i:i + int(seconds * rate)]
 
         # End of signal?
@@ -185,7 +186,7 @@ def specsFromFile(path, rate, seconds, overlap, minlen, shape, start=-1, end=-1,
 
         # Calculate and show noise measure
         noise = signal2noise(spec)
-        print noise
+        print(noise)
 
         # Show spec and wait for enter key
         cv2.imshow('SPEC', spec)

utils/batch_generator.py

@@ -9,6 +9,7 @@
 
 import config as cfg
 from utils import image
+from builtins import range
 
 RANDOM = cfg.getRandomState()
 
@@ -41,7 +42,7 @@ def loadImageAndTarget(sample, augmentation):
 def getDatasetChunk(split):
 
     #get batch-sized chunks of image paths
-    for i in xrange(0, len(split), cfg.BATCH_SIZE):
+    for i in range(0, len(split), cfg.BATCH_SIZE):
         yield split[i:i+cfg.BATCH_SIZE]
 
 def getNextImageBatch(split, augmentation=True): 

utils/log.py

@@ -13,16 +13,16 @@ def show(s, new_line=False):
 
     if isinstance(s, (list, tuple)):
         for i in range(len(s)):
-            print s[i],
+            print(s[i])
             log += str(s[i])
             if i < len(s) - 1:
                 log += ' '
     else:
-        print s,
+        print(s)
         log += str(s)
 
     if new_line:
-        print ''
+        print('')
         log += '\n'
     else:
         log += ' '