Added remove op

theano/typed_list/basic.py

@@ -4,6 +4,8 @@ from theano.gof import Apply, Constant, Op, Variable
 from theano.tensor.type_other import SliceType
 from theano import tensor as T
 
+import numpy
+
 
 class _typed_list_py_operators:
 
@@ -146,7 +148,7 @@ class Insert(Op):
             index = index = T.constant(index, ndim=0)
         else:
             assert isinstance(index, T.TensorVariable) and index.ndim == 0
-        return Apply(self, [x, index, toInsert], [x.ttype()])
+        return Apply(self, [x, index, toInsert], [x.type()])
 
     def perform(self, node, (x, index, toInsert), (out, )):
         if not self.inplace:
@@ -157,3 +159,45 @@ class Insert(Op):
 
     def __str__(self):
         return self.__class__.__name__
+
+
+class Remove(Op):
+
+    def __init__(self, inplace=False):
+        self.inplace = inplace
+        if self.inplace:
+            self.destroy_map = {0: [0]}
+
+    def __eq__(self, other):
+        return type(self) == type(other)
+
+    def __hash__(self):
+        return hash(type(self))
+
+    def make_node(self, x, toRemove):
+        assert isinstance(x.type, TypedListType)
+        assert x.ttype == toRemove.type
+        return Apply(self, [x, toRemove], [x.type()])
+
+    def perform(self, node, (x, toRemove), (out, )):
+
+        if not self.inplace:
+            out[0] = list(x)
+        else:
+            out[0] = x
+
+        """
+        inelegant workaround for ValueError: The truth value of an
+        array with more than one element is ambiguous. Use a.any() or a.all()
+        being thrown when trying to remove a matrix from a matrices list
+        """
+        if isinstance(toRemove, numpy.ndarray):
+            for y in x:
+                if numpy.array_equal(y, toRemove):
+                    toRemove = y
+                    break
+
+        out[0].remove(toRemove)
+
+    def __str__(self):
+        return self.__class__.__name__

theano/typed_list/tests/test_basic.py

@@ -8,7 +8,7 @@ from theano import tensor as T
 from theano.tensor.type_other import SliceType
 from theano.typed_list.type import TypedListType
 from theano.typed_list.basic import (GetItem, Insert,
-                                      Append, Extend)
+                                      Append, Extend, Remove)
 from theano.tests import unittest_tools as utt
 
 
@@ -212,7 +212,6 @@ class test_extend(unittest.TestCase):
 
 class test_insert(unittest.TestCase):
 
-    #FAILING ValueError: expected an ndarray
     def test_inplace(self):
         mySymbolicMatricesList = TypedListType(T.TensorType(
                             theano.config.floatX, (False, False)))()
@@ -245,3 +244,37 @@ class test_insert(unittest.TestCase):
         y = rand_ranged_matrix(-1000, 1000, [100, 101])
 
         self.assertTrue(numpy.array_equal(f([x], numpy.asarray(1), y), [x, y]))
+
+
+class test_remove(unittest.TestCase):
+
+    def test_inplace(self):
+        mySymbolicMatricesList = TypedListType(T.TensorType(
+                            theano.config.floatX, (False, False)))()
+        myMatrix = T.matrix()
+
+        z = Remove(True)(mySymbolicMatricesList, myMatrix)
+
+        f = theano.function([mySymbolicMatricesList, myMatrix], z,
+                            accept_inplace=True)
+
+        x = rand_ranged_matrix(-1000, 1000, [100, 101])
+
+        y = rand_ranged_matrix(-1000, 1000, [100, 101])
+
+        self.assertTrue(numpy.array_equal(f([x, y], y), [x]))
+
+    def test_sanity_check(self):
+        mySymbolicMatricesList = TypedListType(T.TensorType(
+                                theano.config.floatX, (False, False)))()
+        myMatrix = T.matrix()
+
+        z = Remove()(mySymbolicMatricesList, myMatrix)
+
+        f = theano.function([mySymbolicMatricesList, myMatrix], z)
+
+        x = rand_ranged_matrix(-1000, 1000, [100, 101])
+
+        y = rand_ranged_matrix(-1000, 1000, [100, 101])
+
+        self.assertTrue(numpy.array_equal(f([x, y], y), [x]))