Merge pull request #27 from wrznr/fix_gp

Fix the gp module

gramophone/gp/alignment.py

@@ -73,6 +73,9 @@ def align(self,g,p):
         '''
         Aligns a grapheme-phoneme sequence pair.
         '''
+        alignment_fst = self.__align_fst(g,p)
+        if alignment_fst.start() == -1:
+            return []
         return(self.__extract_alignments(self.__align_fst(g,p)))
 
     def scan(self,g):
@@ -93,7 +96,7 @@ def __align_fst(self,g,p):
         t4 = self.expand(p)
         t4.project(project_output=True)
 
-        if t4.num_arcs(t4.start()) == 0:
+        if t4.start() == -1 or t4.num_arcs(t4.start()) == 0:
             return fst.Fst()
 
         t5 = fst.compose(t3,self.E)

gramophone/gp/transcription.py

@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 
 import wapiti
+from tqdm import tqdm
 
 patterns = '''
 # Unigram
@@ -75,7 +76,7 @@ def train(self, training_data):
             self.clear()
 
         # parse training data
-        for alignment in training_data:
+        for alignment in tqdm(training_data):
             seq = u"\n".join(u"%s %s" % (alignment[0][i],alignment[1][i]) for i in range(len(alignment[0])))
             #seq = u"\n".join(u"%s %s" % (x,y) for x,y in zip(alignment[0],alignment[1]))
             self.model.add_training_sequence(seq)

gramophone/scripts/gramophone.py

@@ -92,21 +92,23 @@ def train_gp(mapping,model,data):
     with open(str(data),"r") as f:
         training_data = f.read()
 
-        with click.progressbar(training_data.split("\n")) as bar:
-            for line in bar:
+        for line in tqdm(training_data.split("\n")):
 
-                # skip comments
-                if line.startswith("#"):
-                    continue
+            # skip comments
+            if line.startswith("#"):
+                continue
 
-                # assume tab-separated values
-                fields = line.split("\t")
-                if len(fields) < 2:
-                    continue
+            # assume tab-separated values
+            fields = line.split("\t")
+            if len(fields) < 2:
+                continue
 
-                # align
-                alignment = aligner.align(fields[0],fields[1])
+            # align
+            alignment = aligner.align(fields[0],fields[1])
+            if alignment:
                 aligned_training_data.append(alignment)
+            else:
+                click.echo("%s and %s could not be aligned." % (fields[0], fields[1]))
 
     #
     # stage 2: crf training
@@ -352,7 +354,6 @@ def apply_st(crf,strings):
     # convert
     for string in in_strings:
         encodement = coder.encode(string,mode="scan")
-        click.echo(encodement)
         labellings = labeller.label(encodement)
         combination = []
         for labelling in labellings: