all fixes suggested by Pascal, plus update for the doc

Author: gyom

code/SdA.py

@@ -394,8 +394,7 @@ def test_SdA(finetune_lr=0.1, pretraining_epochs=15,
                 c.append(pretraining_fns[i](index=batch_index,
                          corruption=corruption_levels[i],
                          lr=pretrain_lr))
-            print('Pre-training layer %i, epoch %d, cost ' % (i, epoch))
-            print(numpy.mean(c))
+            print('Pre-training layer %i, epoch %d, cost %f' % (i, epoch, numpy.mean(c)))
 
     end_time = timeit.default_timer()
 

code/lstm.py

@@ -6,7 +6,6 @@
 from six.moves import xrange
 import six.moves.cPickle as pickle
 
-#from six.moves.collections import OrderedDict
 from collections import OrderedDict
 import sys
 import time

code/rnnslu.py

@@ -333,8 +333,8 @@ def main(param=None):
         for i, (x, y) in enumerate(zip(train_lex, train_y)):
             rnn.train(x, y, param['win'], param['clr'])
             print('[learning] epoch %i >> %2.2f%%' % (
-                e, (i + 1) * 100. / nsentences),)
-            print('completed in %.2f (sec) <<\r' % (timeit.default_timer() - tic),)
+                e, (i + 1) * 100. / nsentences), end=' ')
+            print('completed in %.2f (sec) <<\r' % (timeit.default_timer() - tic), end='')
             sys.stdout.flush()
 
         # evaluation // back into the real world : idx -> words
@@ -393,10 +393,10 @@ def main(param=None):
         if param['clr'] < 1e-5:
             break
 
-    print(('BEST RESULT: epoch', param['be'],
+    print('BEST RESULT: epoch', param['be'],
            'valid F1', param['vf1'],
            'best test F1', param['tf1'],
-           'with the model', folder))
+           'with the model', folder)
 
 
 if __name__ == '__main__':

doc/index.txt

@@ -63,3 +63,6 @@ Energy-based recurrent neural network (RNN-RBM):
 .. _Theano basic tutorial: http://deeplearning.net/software/theano/tutorial
 
 .. _Contractive auto-encoders: https://github.com/lisa-lab/DeepLearningTutorials/blob/master/code/cA.py
+
+Note that the tutorials here are all compatible with Python 2 and 3, with the exception of :ref:`rnnrbm` which is only available for Python 2.
+