Merge pull request #229 from pascanur/depricated_scan

theano/scan_module/scan.py

@@ -391,12 +391,11 @@ def scan(fn,
             if isinstance(outs_info[i], dict):
                 # DEPRECATED :
                 if outs_info[i].get('return_steps', None):
-                    _logger.warning(
+                    raise ValueError(
                             "Using `return_steps` has been deprecated. "
                             "Simply select the entries you need using a "
                             "subtensor. Scan will optimize memory "
                             "consumption, so do not worry about that.")
-                    return_steps[i] = outs_info[i]['return_steps']
                 # END
 
             if not isinstance(outs_info[i], dict):

theano/scan_module/scan_utils.py

@@ -208,8 +208,7 @@ def get_updates_and_outputs(ls):
                 raise ValueError(error_msg)
         elif is_updates(ls[0]):
             if is_outputs(ls[1]):
-                _logger.warning(deprication_msg)
-                return (None, _list(ls[1]), dict(ls[0]))
+                raise ValueError(deprication_msg)
             elif is_condition(ls[1]):
                 return (ls[1].condition, [], dict(ls[0]))
             else:
@@ -225,15 +224,6 @@ def get_updates_and_outputs(ls):
                     raise ValueError(error_msg)
             else:
                 raise ValueError(error_msg)
-        elif is_updates(ls[0]):
-            if is_outputs(ls[1]):
-                if is_condition(ls[2]):
-                    _logger.warning(deprication_msg)
-                    return (ls[2].condition, _list(ls[1]), dict(ls[0]))
-                else:
-                    raise ValueError(error_msg)
-            else:
-                raise ValueError(error_msg)
         else:
             raise ValueError(error_msg)
 

theano/scan_module/tests/test_scan.py

@@ -2560,18 +2560,21 @@ class T_Scan(unittest.TestCase):
                     theano.dot(x_tm1, W),
                     y_tm1 + theano.dot(x_tm1, W_out)]
 
-        outputs, updates = theano.scan( f_rnn_cmpl
+        rval, updates = theano.scan( f_rnn_cmpl
                                        , [ u1
                                           , u2]
-                                       , [ dict(store_steps = 3)
-                                          , dict(initial = x0, return_steps = 2)
-                                          , dict(initial=y0, taps=[-1,-3],
-                                                 return_steps = 4)]
+                                       , [ None
+                                          , dict(initial = x0)
+                                          , dict(initial=y0, taps=[-1,-3])]
                                        , W_in1
                                        , n_steps           = None
                                        , truncate_gradient = -1
                                        , go_backwards      = False)
 
+        outputs = []
+        outputs += [rval[0][-3:]]
+        outputs += [rval[1][-2:]]
+        outputs += [rval[2][-4:]]
         f4     = theano.function([u1,u2,x0,y0,W_in1], outputs
                                  , updates = updates
                                  , allow_input_downcast = True