Fix Python 3 syntax errors

bleu/bleu.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# 
+#
 # File Name : bleu.py
 #
 # Description : Wrapper for BLEU scorer.
@@ -8,7 +8,7 @@
 # Last Modified : Thu 19 Mar 2015 09:13:28 PM PDT
 # Authors : Hao Fang <[email protected]> and Tsung-Yi Lin <[email protected]>
 
-from bleu_scorer import BleuScorer
+from .bleu_scorer import BleuScorer
 
 
 class Bleu:

bleu/bleu_scorer.py

@@ -7,7 +7,7 @@
 # reserved. Do not redistribute without permission from the
 # author. Not for commercial use.
 
-# Modified by: 
+# Modified by:
 # Hao Fang <[email protected]>
 # Tsung-Yi Lin <[email protected]>
 
@@ -52,21 +52,22 @@ def cook_refs(refs, eff=None, n=4): ## lhuang: oracle will call with "average"
         reflen = float(sum(reflen))/len(reflen)
 
     ## lhuang: N.B.: leave reflen computaiton to the very end!!
-    
+
     ## lhuang: N.B.: in case of "closest", keep a list of reflens!! (bad design)
 
     return (reflen, maxcounts)
 
-def cook_test(test, (reflen, refmaxcounts), eff=None, n=4):
+def cook_test(test, refs, eff=None, n=4):
     '''Takes a test sentence and returns an object that
     encapsulates everything that BLEU needs to know about it.'''
 
+    reflen, refmaxcounts = refs
     testlen, counts = precook(test, n, True)
 
     result = {}
 
     # Calculate effective reference sentence length.
-    
+
     if eff == "closest":
         result["reflen"] = min((abs(l-testlen), l) for l in reflen)[1]
     else: ## i.e., "average" or "shortest" or None
@@ -108,7 +109,7 @@ def __init__(self, test=None, refs=None, n=4, special_reflen=None):
 
     def cook_append(self, test, refs):
         '''called by constructor and __iadd__ to avoid creating new instances.'''
-        
+
         if refs is not None:
             self.crefs.append(cook_refs(refs))
             if test is not None:
@@ -136,7 +137,7 @@ def reflen(self, option=None):
 
     def testlen(self, option=None):
         self.compute_score(option=option)
-        return self._testlen        
+        return self._testlen
 
     def retest(self, new_test):
         if type(new_test) is str:
@@ -151,7 +152,7 @@ def retest(self, new_test):
 
     def rescore(self, new_test):
         ''' replace test(s) with new test(s), and returns the new score.'''
-        
+
         return self.retest(new_test).compute_score()
 
     def size(self):
@@ -170,7 +171,7 @@ def __iadd__(self, other):
             self.crefs.extend(other.crefs)
             self._score = None ## need to recompute
 
-        return self        
+        return self
 
     def compatible(self, other):
         return isinstance(other, BleuScorer) and self.n == other.n
@@ -179,7 +180,7 @@ def single_reflen(self, option="average"):
         return self._single_reflen(self.crefs[0][0], option)
 
     def _single_reflen(self, reflens, option=None, testlen=None):
-        
+
         if option == "shortest":
             reflen = min(reflens)
         elif option == "average":
@@ -194,7 +195,7 @@ def _single_reflen(self, reflens, option=None, testlen=None):
     def recompute_score(self, option=None, verbose=0):
         self._score = None
         return self.compute_score(option, verbose)
-        
+
     def compute_score(self, option=None, verbose=0):
         n = self.n
         small = 1e-9
@@ -212,7 +213,7 @@ def compute_score(self, option=None, verbose=0):
         totalcomps = {'testlen':0, 'reflen':0, 'guess':[0]*n, 'correct':[0]*n}
 
         # for each sentence
-        for comps in self.ctest:            
+        for comps in self.ctest:
             testlen = comps['testlen']
             self._testlen += testlen
 
@@ -222,7 +223,7 @@ def compute_score(self, option=None, verbose=0):
                 reflen = self.special_reflen
 
             self._reflen += reflen
-                
+
             for key in ['guess','correct']:
                 for k in xrange(n):
                     totalcomps[key][k] += comps[key][k]
@@ -231,15 +232,15 @@ def compute_score(self, option=None, verbose=0):
             bleu = 1.
             for k in xrange(n):
                 bleu *= (float(comps['correct'][k]) + tiny) \
-                        /(float(comps['guess'][k]) + small) 
+                        /(float(comps['guess'][k]) + small)
                 bleu_list[k].append(bleu ** (1./(k+1)))
             ratio = (testlen + tiny) / (reflen + small) ## N.B.: avoid zero division
             if ratio < 1:
                 for k in xrange(n):
                     bleu_list[k][-1] *= math.exp(1 - 1/ratio)
 
             if verbose > 1:
-                print comps, reflen
+                print(comps, reflen)
 
         totalcomps['reflen'] = self._reflen
         totalcomps['testlen'] = self._testlen
@@ -256,8 +257,8 @@ def compute_score(self, option=None, verbose=0):
                 bleus[k] *= math.exp(1 - 1/ratio)
 
         if verbose > 0:
-            print totalcomps
-            print "ratio:", ratio
+            print(totalcomps)
+            print("ratio:", ratio)
 
         self._score = bleus
         return self._score, bleu_list

cider/cider.py

@@ -1,18 +1,18 @@
 # Filename: cider.py
 #
-# Description: Describes the class to compute the CIDEr (Consensus-Based Image Description Evaluation) Metric 
+# Description: Describes the class to compute the CIDEr (Consensus-Based Image Description Evaluation) Metric
 #               by Vedantam, Zitnick, and Parikh (http://arxiv.org/abs/1411.5726)
 #
 # Creation Date: Sun Feb  8 14:16:54 2015
 #
 # Authors: Ramakrishna Vedantam <[email protected]> and Tsung-Yi Lin <[email protected]>
 
-from cider_scorer import CiderScorer
+from .cider_scorer import CiderScorer
 import pdb
 
 class Cider:
     """
-    Main Class to compute the CIDEr metric 
+    Main Class to compute the CIDEr metric
 
     """
     def __init__(self, test=None, refs=None, n=4, sigma=6.0):
@@ -26,7 +26,7 @@ def compute_score(self, gts, res):
         Main function to compute CIDEr score
         :param  hypo_for_image (dict) : dictionary with key <image> and value <tokenized hypothesis / candidate sentence>
                 ref_for_image (dict)  : dictionary with key <image> and value <tokenized reference sentence>
-        :return: cider (float) : computed CIDEr score for the corpus 
+        :return: cider (float) : computed CIDEr score for the corpus
         """
 
         assert(gts.keys() == res.keys())
@@ -51,4 +51,4 @@ def compute_score(self, gts, res):
         return score, scores
 
     def method(self):
-        return "CIDEr"
+        return "CIDEr"

eval.py

@@ -1,9 +1,9 @@
 __author__ = 'tylin'
-from tokenizer.ptbtokenizer import PTBTokenizer
-from bleu.bleu import Bleu
-from meteor.meteor import Meteor
-from rouge.rouge import Rouge
-from cider.cider import Cider
+from .tokenizer.ptbtokenizer import PTBTokenizer
+from .bleu.bleu import Bleu
+from .meteor.meteor import Meteor
+from .rouge.rouge import Rouge
+from .cider.cider import Cider
 
 class COCOEvalCap:
     def __init__(self, coco, cocoRes):
@@ -26,15 +26,15 @@ def evaluate(self):
         # =================================================
         # Set up scorers
         # =================================================
-        print 'tokenization...'
+        print('tokenization...')
         tokenizer = PTBTokenizer()
         gts  = tokenizer.tokenize(gts)
         res = tokenizer.tokenize(res)
 
         # =================================================
         # Set up scorers
         # =================================================
-        print 'setting up scorers...'
+        print('setting up scorers...')
         scorers = [
             (Bleu(4), ["Bleu_1", "Bleu_2", "Bleu_3", "Bleu_4"]),
             (Meteor(),"METEOR"),
@@ -46,17 +46,17 @@ def evaluate(self):
         # Compute scores
         # =================================================
         for scorer, method in scorers:
-            print 'computing %s score...'%(scorer.method())
+            print('computing %s score...'%(scorer.method()))
             score, scores = scorer.compute_score(gts, res)
             if type(method) == list:
                 for sc, scs, m in zip(score, scores, method):
                     self.setEval(sc, m)
                     self.setImgToEvalImgs(scs, gts.keys(), m)
-                    print "%s: %0.3f"%(m, sc)
+                    print("%s: %0.3f"%(m, sc))
             else:
                 self.setEval(score, method)
                 self.setImgToEvalImgs(scores, gts.keys(), method)
-                print "%s: %0.3f"%(method, score)
+                print("%s: %0.3f"%(method, score))
         self.setEvalImgs()
 
     def setEval(self, score, method):
@@ -70,4 +70,4 @@ def setImgToEvalImgs(self, scores, imgIds, method):
             self.imgToEval[imgId][method] = score
 
     def setEvalImgs(self):
-        self.evalImgs = [eval for imgId, eval in self.imgToEval.items()]
+        self.evalImgs = [eval for imgId, eval in self.imgToEval.items()]

meteor/meteor.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python
 
 # Python wrapper for METEOR implementation, by Xinlei Chen
-# Acknowledge Michael Denkowski for the generous discussion and help 
+# Acknowledge Michael Denkowski for the generous discussion and help
 
 import os
 import sys
@@ -63,15 +63,15 @@ def _score(self, hypothesis_str, reference_list):
         self.meteor_p.stdin.write('{}\n'.format(score_line))
         stats = self.meteor_p.stdout.readline().strip()
         eval_line = 'EVAL ||| {}'.format(stats)
-        # EVAL ||| stats 
+        # EVAL ||| stats
         self.meteor_p.stdin.write('{}\n'.format(eval_line))
         score = float(self.meteor_p.stdout.readline().strip())
         # bug fix: there are two values returned by the jar file, one average, and one all, so do it twice
         # thanks for Andrej for pointing this out
         score = float(self.meteor_p.stdout.readline().strip())
         self.lock.release()
         return score
- 
+
     def __del__(self):
         self.lock.acquire()
         self.meteor_p.stdin.close()

rouge/rouge.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# 
+#
 # File Name : rouge.py
 #
 # Description : Computes ROUGE-L metric as described by Lin and Hovey (2004)
@@ -49,14 +49,14 @@ def calc_score(self, candidate, refs):
         :param refs: list of str : COCO reference sentences for the particular image to be evaluated
         :returns score: int (ROUGE-L score for the candidate evaluated against references)
         """
-        assert(len(candidate)==1)	
-        assert(len(refs)>0)         
+        assert(len(candidate)==1)
+        assert(len(refs)>0)
         prec = []
         rec = []
 
         # split into tokens
         token_c = candidate[0].split(" ")
-    	
+
         for reference in refs:
             # split into tokens
             token_r = reference.split(" ")
@@ -77,8 +77,8 @@ def calc_score(self, candidate, refs):
     def compute_score(self, gts, res):
         """
         Computes Rouge-L score given a set of reference and candidate sentences for the dataset
-        Invoked by evaluate_captions.py 
-        :param hypo_for_image: dict : candidate / test sentences with "image name" key and "tokenized sentences" as values 
+        Invoked by evaluate_captions.py
+        :param hypo_for_image: dict : candidate / test sentences with "image name" key and "tokenized sentences" as values
         :param ref_for_image: dict : reference MS-COCO sentences with "image name" key and "tokenized sentences" as values
         :returns: average_score: float (mean ROUGE-L score computed by averaging scores for all the images)
         """

tokenizer/ptbtokenizer.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python
-# 
+#
 # File Name : ptbtokenizer.py
 #
 # Description : Do the PTB Tokenization and remove punctuations.
@@ -19,7 +19,7 @@
 
 # punctuations to be removed from the sentences
 PUNCTUATIONS = ["''", "'", "``", "`", "-LRB-", "-RRB-", "-LCB-", "-RCB-", \
-        ".", "?", "!", ",", ":", "-", "--", "...", ";"] 
+        ".", "?", "!", ",", ":", "-", "--", "...", ";"]
 
 class PTBTokenizer:
     """Python wrapper of Stanford PTBTokenizer"""