fix is tensor test

contrib/Python/cntk/ops/__init__.py

@@ -940,7 +940,7 @@ def input_numpy(value, alias=None, dynamic_axis='', name=None):
         :class:`cntk.graph.ComputationNode`
     '''
     from .. import utils
-    if utils.is_tensor_list(value) or utils.is_tensor(value):
+    if utils.is_tensor(value) or utils.is_tensor(value):
         value = np.asarray(value)
         if dynamic_axis:
             cntk_shape = value[0].shape[1:]

contrib/Python/cntk/utils/__init__.py

@@ -189,7 +189,7 @@ def is_tensor(data):
         if isinstance(data, np.ndarray):
             return True
 
-        if not isinstance(data[0], list):
+        if not (isinstance(data[0], list) or isinstance(data[0], np.ndarray)):
             return False
 
         data = data[0]

contrib/Python/cntk/utils/eval.py

@@ -43,12 +43,8 @@ def eval(node):
                         # One param needs to be an Input() node. This will be fixed in 
                         # CNTK soon, so that we can remove this workaround and evaluate a 
                         # network with no inputs.
-                        if first:
-                            if not isinstance(val, list):                
-                                # inputs have the outmost dimension for sequence dimension
-                                val = [val]
-        
-                            ir = input_numpy(val, alias=p, name=p)
+                        if first:        
+                            ir = input_numpy([val], alias=p, name=p)
                             setattr(node, p, ir)
                             first = False
                         else:
@@ -56,5 +52,5 @@ def eval(node):
                     else:
                         if isinstance(val, _InputComputationNodeBase) and first:
                             first = False
-        ctx.clean_up=False                   
+                          
         return ctx.eval(node)