py3k: Use print function instead of statement

OLD/ocrolib/h5utils.py

@@ -1,3 +1,5 @@
+from __future__ import print_function
+
 from numpy import dtype
 import tables
 import re
@@ -10,7 +12,7 @@ def table_copy(source,dest,names=None,omit=[],verbose=1):
         names = names.split()
     for name in names:
         a = source.getNode("/"+name)
-        if verbose: print "[copying",name,a.shape,a.atom,"]"
+        if verbose: print("[copying", name, a.shape, a.atom, "]")
         if "VLArray" in str(a):
             b = dest.createVLArray(dest.root,name,a.atom,filters=tables.Filters(9))
         else:
@@ -28,7 +30,7 @@ def assign_array(db,name,a,verbose=1):
         atom = tables.Float32Atom()
     else:
         raise Exception('unknown array type: %s'%a.dtype)
-    if verbose: print "[writing",name,a.shape,atom,"]"
+    if verbose: print("[writing", name, a.shape, atom, "]")
     node = db.createEArray(db.root,name,atom,shape=[0]+list(a.shape[1:]),filters=tables.Filters(9))
     node.append(a)
 

OLD/ocrolib/ngraphs.py

@@ -1,6 +1,8 @@
 # TODO:
 # - handle UTF-8 inputs correctly
 
+from __future__ import print_function
+
 from pylab import *
 from collections import Counter,defaultdict
 import re
@@ -47,7 +49,7 @@ def safe_readlines(stream,nonl=0):
         try:
             line = stream.readline()
         except UnicodeDecodeError as e:
-            if not once: print lineno,":",e
+            if not once: print(lineno, ":", e)
             once = 1
             return
         if line is None: return
@@ -75,14 +77,14 @@ def computeNGraphs(self,fnames,n):
         lineskip = 0
         linelimit = 2000
         for fnum,fname in enumerate(fnames):
-            print fnum,"of",len(fnames),":",fname
+            print(fnum, "of", len(fnames), ":", fname)
             if fname.startswith("lineskip="):
                 lineskip = int(fname.split("=")[1])
-                print "changing lineskip to",lineskip
+                print("changing lineskip to", lineskip)
                 continue
             if fname.startswith("linelimit="):
                 linelimit = int(fname.split("=")[1])
-                print "changing linelimit to",linelimit
+                print("changing linelimit to", linelimit)
                 continue
             with codecs.open(fname,"r","utf-8") as stream:
                 for lineno,line in enumerate(safe_readlines(stream)):
@@ -106,11 +108,11 @@ def __init__(self,*args,**kw):
         NGraphsCounts.__init__(self,*args,**kw)
     def buildFromFiles(self,fnames,n):
         """Given a set of files, build the log posteriors."""
-        print "reading",len(fnames),"files"
+        print("reading", len(fnames), "files")
         counter = self.computeNGraphs(fnames,n)
-        print "got",sum(counter.values()),"%d-graphs"%(n,)
+        print("got", sum(counter.values()), "%d-graphs" % (n, ))
         self.computePosteriors(counter)
-        print "done building lposteriors"
+        print("done building lposteriors")
     def computePosteriors(self,counter):
         """Given a `counter` of all n-graphs, compute
         (log) conditional probabilities."""

examples/rnnlib-vs-ocropy.ipynb

@@ -21,6 +21,8 @@
      "cell_type": "code",
      "collapsed": false,
      "input": [
+      "from future import print_function\n",
+      "\n",
       "from pylab import *\n",
       "import os.path\n",
       "import glob\n",
@@ -60,7 +62,7 @@
      "cell_type": "code",
      "collapsed": false,
      "input": [
-      "print \"c_lstm\"\n",
+      "print(\"c_lstm\")\n",
       "clstm = lstm_c.C_LSTM(Ni,Ns,No)\n",
       "llayer = rnnscript.as_lstm2(clstm.net.hiddenLayers[0])\n",
       "clstm.setLearningRate(1e-4,0.9)\n",
@@ -72,7 +74,7 @@
       "    err = sum((a-ys)**2)**.5\n",
       "    errs.append(err)\n",
       "    if i%10000==0:\n",
-      "        print i,mean(errs[-1000:])\n",
+      "        print(i, mean(errs[-1000:]))\n",
       "        s = clstm.states()\n",
       "        clf()\n",
       "        subplot(121);\n",
@@ -83,7 +85,7 @@
       "        draw()\n",
       "\n",
       "test_xs,test_ys = genseq()\n",
-      "print \"done\""
+      "print(\"done\")"
      ],
      "language": "python",
      "metadata": {},
@@ -221,7 +223,7 @@
       "plot(deltas3,color='green',linestyle='dashed')\n",
       "deltas4 = -array(before.nets[1].dzspre)\n",
       "plot(deltas4,color='black',linestyle='dashed')\n",
-      "print deltas4.shape"
+      "print(deltas4.shape)"
      ],
      "language": "python",
      "metadata": {},
@@ -289,12 +291,12 @@
       "clstm.ctrain(xs,ys,update=0)\n",
       "h2o.debug = False\n",
       "b2o.debug = False\n",
-      "print\n",
-      "print array(h2o.dinputs).reshape(3,2)\n",
-      "print array(h2o.dweights).reshape(3,2)\n",
-      "print\n",
-      "print array(b2o.dinputs).reshape(3)\n",
-      "print array(b2o.dweights).reshape(3)"
+      "print()\n",
+      "print(array(h2o.dinputs).reshape(3, 2))\n",
+      "print(array(h2o.dweights).reshape(3, 2))\n",
+      "print()\n",
+      "print(array(b2o.dinputs).reshape(3))\n",
+      "print(array(b2o.dweights).reshape(3))"
      ],
      "language": "python",
      "metadata": {},
@@ -364,4 +366,4 @@
    "metadata": {}
   }
  ]
-}
+}

ocrolib/common.py

@@ -3,6 +3,8 @@
 ### common functions for data structures, file name manipulation, etc.
 ################################################################
 
+from __future__ import print_function
+
 import os
 import os.path
 import re
@@ -96,7 +98,7 @@ def _wrap(f):
         warned = 0
         def _wrapper(*args,**kw):
             if not warned:
-                print f,"has been DEPRECATED"
+                print(f, "has been DEPRECATED")
                 warned = 1
             return f(*args,**kw)
     return _wrap
@@ -248,7 +250,7 @@ def write_image_gray(fname,image,normalize=0,verbose=0):
     type, its values are clipped to the range [0,1],
     multiplied by 255 and converted to unsigned bytes.  Otherwise,
     the image must be of type unsigned byte."""
-    if verbose: print "# writing",fname
+    if verbose: print("# writing", fname)
     if isfloatarray(image):
         image = array(255*clip(image,0.0,1.0),'B')
     assert image.dtype==dtype('B'),"array has wrong dtype: %s"%image.dtype
@@ -271,7 +273,7 @@ def write_image_binary(fname,image,verbose=0):
     """Write a binary image to disk. This verifies first that the given image
     is, in fact, binary.  The image may be of any type, but must consist of only
     two values."""
-    if verbose: print "# writing",fname
+    if verbose: print("# writing", fname)
     assert image.ndim==2
     image = array(255*(image>midrange(image)),'B')
     im = array2pil(image)
@@ -428,7 +430,7 @@ def bbox(self,i):
         """Return the bounding box in raster coordinates
         (row0,col0,row1,col1)."""
         r = self.objects[i]
-        # print "@@@bbox",i,r
+        # print("@@@bbox", i, r)
         return (r[0].start,r[1].start,r[0].stop,r[1].stop)
     def bboxMath(self,i):
         """Return the bounding box in math coordinates
@@ -442,7 +444,7 @@ def length(self):
     def mask(self,index,margin=0):
         """Return the mask for component index."""
         b = self.objects[index]
-        #print "@@@mask",index,b
+        # print("@@@mask", index, b)
         m = self.labels[b]
         m[m!=index] = 0
         if margin>0: m = pad_by(m,margin)
@@ -499,7 +501,7 @@ class names that have changed."""
     if not nofind:
         fname = ocropus_find_file(fname)
     if verbose:
-        print "# loading object",fname
+        print("# loading object", fname)
     if zip==0 and fname.endswith(".gz"):
         zip = 1
     if zip>0:
@@ -833,7 +835,7 @@ def pyconstruct(s):
     path = s[:s.find("(")]
     if "." in path:
         module = path[:path.rfind(".")]
-        print "import",module
+        print("import", module)
         exec "import "+module in env
     return eval(s,env)
 
@@ -890,15 +892,15 @@ def save_component(file,object,verbose=0,verify=0):
         ocropus.save_component(file,object)
         return
     if verbose:
-        print "[save_component]"
+        print("[save_component]")
     if verbose:
         for k,v in object.__dict__.items():
-            print ":",k,obinfo(v)
+            print(":", k, obinfo(v))
     with open(file,"wb") as stream:
         pickle.dump(object,stream,pickle_mode)
     if verify:
         if verbose:
-            print "[trying to read it again]"
+            print("[trying to read it again]")
         with open(file,"rb") as stream:
             pickle.load(stream)
 

ocrolib/edist.py

@@ -72,8 +72,8 @@ def xlevenshtein(a,b,context=1):
     same = filters.minimum_filter(same,1+2*context)
     als = "".join([al[i] if not same[i] else "~" for i in range(len(al))])
     bls = "".join([bl[i] if not same[i] else "~" for i in range(len(bl))])
-    # print als
-    # print bls
+    # print(als)
+    # print(bls)
     ags = re.split(r'~+',als)
     bgs = re.split(r'~+',bls)
     confusions = [(a,b) for a,b in zip(ags,bgs) if a!="" or b!=""]

ocrolib/extras/fgen.py

@@ -2,6 +2,8 @@
 ### text image generation with Cairo
 ################################################################
 
+from __future__ import print_function
+
 import ctypes
 import cairo
 from cairoextras import *
@@ -104,7 +106,7 @@ def pango_render_string(s,spec=None,fontfile=None,size=None,bg=(0.0,0.0,0.0),fg=
         else:
             layout.set_markup(s)
         ((xbear,ybear,tw,th),_) = layout.get_pixel_extents()
-        # print xbear,ybear,tw,th
+        # print(xbear, ybear, tw, th)
         tw = tw+2*pad
         th = th+2*pad
         if tw<=w and th<=h: break
@@ -160,12 +162,12 @@ def gauss_distort(images,maxdelta=2.0,sigma=10.0):
     deltas /= max(amax(deltas),-amin(deltas))
     deltas *= maxdelta
     xy = transpose(array(meshgrid(range(n),range(m))),axes=[0,2,1])
-    # print xy.shape,deltas.shape
+    # print(xy.shape, deltas.shape)
     deltas +=  xy
     return [map_coordinates(image,deltas,order=1) for image in images]
 
 if __name__=="__main__":
-    # print sorted(pango_families())
+    # print(sorted(pango_families()))
     ion()
     show()
     while 1:

ocrolib/extras/lru.py

@@ -3,6 +3,8 @@
 # Source: http://code.activestate.com/recipes/498245-lru-and-lfu-cache-decorators/
 # Author: Raymond Hettinger
 
+from __future__ import print_function
+
 import collections
 import functools
 from itertools import ifilterfalse

ocrolib/lineest.py

@@ -1,3 +1,5 @@
+from __future__ import print_function
+
 import sys
 import os
 import re
@@ -21,7 +23,7 @@ def __init__(self,target_height=48,params=(4,1.0,0.3)):
         self.debug = int(os.getenv("debug_center") or "0")
         self.target_height = target_height
         self.range,self.smoothness,self.extra = params
-        print "# CenterNormalizer"
+        print("# CenterNormalizer")
     def setHeight(self,target_height):
         self.target_height = target_height
     def measure(self,line):

ocrolib/lstm.py

@@ -24,6 +24,8 @@
 # Author: Thomas M. Breuel
 # License: Apache 2.0
 
+from __future__ import print_function
+
 import common as ocrolib
 from pylab import *
 from collections import defaultdict
@@ -223,7 +225,7 @@ def update(self):
             ds.ravel()[:] = self.momentum * ds.ravel()[:] + self.learning_rate * dw.ravel()[:]
             w.ravel()[:] += ds.ravel()[:]
             if self.verbose:
-                print n,(amin(w),amax(w)),(amin(dw),amax(dw))
+                print(n, (amin(w), amax(w)), (amin(dw), amax(dw)))
 
 ''' The following are subclass responsibility:
 
@@ -281,7 +283,7 @@ def info(self):
         vars = sorted("W2".split())
         for v in vars:
             a = array(getattr(self,v))
-            print v,a.shape,amin(a),amax(a)
+            print(v, a.shape, amin(a), amax(a))
     def weights(self):
         yield self.W2,self.DW2,"Logreg"
 
@@ -325,7 +327,7 @@ def info(self):
         vars = sorted("W2".split())
         for v in vars:
             a = array(getattr(self,v))
-            print v,a.shape,amin(a),amax(a)
+            print(v, a.shape, amin(a), amax(a))
     def weights(self):
         yield self.W2,self.DW2,"Softmax"
 
@@ -514,7 +516,7 @@ def info(self):
         vars = sorted(vars)
         for v in vars:
             a = array(getattr(self,v))
-            print v,a.shape,amin(a),amax(a)
+            print(v, a.shape, amin(a), amax(a))
     def preSave(self):
         self.max_n = max(500,len(self.ci))
         self.allocate(1)

ocrolib/morph.py

@@ -2,6 +2,8 @@
 ### various add-ons to the SciPy morphology package
 ################################################################
 
+from __future__ import print_function
+
 from numpy import *
 import pylab
 from pylab import *
@@ -198,9 +200,9 @@ def select_regions(binary,f,min=0,nbest=100000):
     for i in best[-nbest:]:
         if scores[i]<=min: continue
         keep[i+1] = 1
-    # print scores,best[-nbest:],keep
-    # print sorted(list(set(labels.ravel())))
-    # print sorted(list(set(keep[labels].ravel())))
+    # print(scores, best[-nbest:], keep)
+    # print(sorted(list(set(labels.ravel()))))
+    # print(sorted(list(set(keep[labels].ravel()))))
     return keep[labels]
 
 @checks(SEGMENTATION)