End & array_equal & shape

pyttb/cp_als.py

@@ -197,7 +197,7 @@ def cp_als(
             # Compute the matrix of coefficients for linear system
             Y = np.prod(UtU, axis=2, where=[i != n for i in range(N)])
             # don't try to solve linear system with Y = 0
-            if (Y == np.zeros(Y.shape)).all():
+            if (Y == 0).all():
                 Unew = np.zeros(Unew.shape)
             else:
                 Unew = np.linalg.solve(Y.T, Unew.T).T
@@ -213,7 +213,7 @@ def cp_als(
                 weights = np.maximum(np.max(np.abs(Unew), 0), 1)  # max-norm
 
             # if weights are 0, do not divide
-            if not (weights == np.zeros(weights.shape)).all():
+            if not (weights == 0).all():
                 Unew = Unew / weights
 
             U[n] = Unew

pyttb/ktensor.py

@@ -614,31 +614,6 @@ def double(self) -> np.ndarray:
         """
         return self.full().double()
 
-    def end(self, k: Optional[int] = None) -> int:
-        """
-        Last index of indexing expression for :class:`pyttb.ktensor`.
-
-        Parameters
-        ----------
-        k:
-            Dimension for subscripted indexing
-
-        Returns
-        -------
-        Final index
-
-        Examples
-        --------
-        >>> K = ttb.ktensor.from_function(np.ones, (2, 3, 4), 2)
-        >>> print(K.end(2))
-        3
-        """
-
-        if k is not None:  # Subscripted indexing
-            return self.shape[k] - 1
-        # For linear indexing
-        return int(np.prod(self.shape) - 1)
-
     def extract(
         self, idx: Optional[Union[int, tuple, list, np.ndarray]] = None
     ) -> ktensor:
@@ -893,7 +868,7 @@ def full(self) -> ttb.tensor:
         [[5. 6.]
          [7. 8.]]
         >>> print(K.full()) # doctest: +NORMALIZE_WHITESPACE
-        tensor of shape 2 x 2
+        tensor of shape (2, 2)
         data[:, :] =
         [[29. 39.]
          [63. 85.]]
@@ -976,7 +951,7 @@ def isequal(self, other: ttb.ktensor) -> bool:
         if (self.weights != other.weights).any():
             return False
         for k in range(self.ndims):
-            if not (self.factor_matrices[k] == other.factor_matrices[k]).all():
+            if not np.array_equal(self.factor_matrices[k], other.factor_matrices[k]):
                 return False
         return True
 
@@ -1032,7 +1007,7 @@ def issymmetric(
             for j in range(i + 1, self.ndims):
                 if self.factor_matrices[i].shape != self.factor_matrices[j].shape:
                     diffs[i, j] = np.inf
-                elif (self.factor_matrices[i] == self.factor_matrices[j]).all():
+                elif np.array_equal(self.factor_matrices[i], self.factor_matrices[j]):
                     diffs[i, j] = 0
                 else:
                     diffs[i, j] = np.linalg.norm(
@@ -1771,7 +1746,7 @@ def tolist(self, mode: Optional[int] = None) -> List[np.ndarray]:
             assert False, "Input parameter'mode' must be in the range of self.ndims"
 
         # all weights are equal to 1
-        if (self.weights == np.ones(self.weights.shape)).all():
+        if np.array_equal(self.weights, np.ones(self.weights.shape)):
             return self.factor_matrices.copy()
 
         lsgn = np.sign(self.weights)

pyttb/sptensor.py

@@ -429,7 +429,7 @@ def collapse(
                     size=newsize[0],
                     func=fun,
                 )
-            return np.zeros((newsize[0], 1))
+            return np.zeros((newsize[0],))
 
         # Create Result
         if self.subs.size > 0:
@@ -533,19 +533,6 @@ def elemfun(self, function_handle: Callable[[np.ndarray], np.ndarray]) -> sptens
             return ttb.sptensor.from_data(np.array([]), np.array([]), self.shape)
         return ttb.sptensor.from_data(self.subs[idx, :], vals[idx], self.shape)
 
-    def end(self, k: Optional[int] = None) -> int:
-        """
-        Last index of indexing expression for sparse tensor
-
-        Parameters
-        ----------
-        k:
-            Dimension for subscript indexing
-        """
-        if k is not None:
-            return self.shape[k] - 1
-        return int(np.prod(self.shape) - 1)
-
     def extract(self, searchsubs: np.ndarray) -> np.ndarray:
         """
         Extract value for a sptensor.
@@ -1102,7 +1089,7 @@ def scale(self, factor: np.ndarray, dims: Union[float, np.ndarray]) -> sptensor:
                 assert False, "Size mismatch in scale"
             return ttb.sptensor.from_data(
                 self.subs,
-                self.vals * factor[self.subs[:, dims].transpose()[0]],
+                self.vals * factor[self.subs[:, dims].transpose()[0]][:, None],
                 self.shape,
             )
         assert False, "Invalid scaling factor"
@@ -2553,8 +2540,7 @@ def __repr__(self):  # pragma: no cover
             s += (" x ").join([str(int(d)) for d in self.shape])
             return s
 
-        s = "Sparse tensor of shape "
-        s += (" x ").join([str(int(d)) for d in self.shape])
+        s = f"Sparse tensor of shape {self.shape}"
         s += f" with {nz} nonzeros \n"
 
         # Stop insane printouts

pyttb/tenmat.py

@@ -243,22 +243,6 @@ def double(self) -> np.ndarray:
         """
         return self.data.astype(np.float_).copy()
 
-    def end(self, k: int) -> int:
-        """
-        Last index of indexing expression for tenmat
-
-        Parameters
-        ----------
-        k:
-            Dimension for subscripted indexing
-
-        Returns
-        -------
-        Index
-        """
-
-        return self.data.shape[k] - 1
-
     @property
     def ndims(self) -> int:
         """Return the number of dimensions of a tenmat"""
@@ -498,10 +482,7 @@ def __repr__(self):
         """
         s = ""
         s += "matrix corresponding to a tensor of shape "
-        if self.data.size == 0:
-            s += str(self.shape)
-        else:
-            s += (" x ").join([str(int(d)) for d in self.tshape])
+        s += str(self.tshape)
         s += "\n"
 
         s += "rindices = "

pyttb/tensor.py

@@ -335,29 +335,6 @@ def exp(self) -> tensor:
         """
         return ttb.tensor.from_data(np.exp(self.data))
 
-    def end(self, k: Optional[int] = None) -> int:
-        """
-        Last index of indexing expression for tensor
-
-        Parameters
-        ----------
-        k:
-            Dimension for subscripted indexing
-
-        Examples
-        --------
-        >>> X = ttb.tensor.from_data(np.ones((2,2)))
-        >>> X.end()  # linear indexing
-        3
-        >>> X.end(0)
-        1
-        """
-
-        if k is not None:  # Subscripted indexing
-            return self.shape[k] - 1
-        # For linear indexing
-        return np.prod(self.shape) - 1
-
     def find(self) -> Tuple[np.ndarray, np.ndarray]:
         """
         FIND Find subscripts of nonzero elements in a tensor.
@@ -1462,12 +1439,12 @@ def __getitem__(self, item):
         1.0
         >>> # produces a tensor of order 1 and size 1
         >>> X[1,1,1,:] # doctest: +NORMALIZE_WHITESPACE
-        tensor of shape 1
+        tensor of shape (1,)
         data[:] =
         [1.]
         >>> # produces a tensor of size 2 x 2 x 1
         >>> X[0:2,[2, 3],1,:] # doctest: +NORMALIZE_WHITESPACE
-        tensor of shape 2 x 2 x 1
+        tensor of shape (2, 2, 1)
         data[0, :, :] =
         [[1.]
          [1.]]
@@ -1862,8 +1839,7 @@ def __repr__(self):
             return s
 
         s = ""
-        s += "tensor of shape "
-        s += (" x ").join([str(int(d)) for d in self.shape])
+        s += f"tensor of shape {self.shape}"
         s += "\n"
 
         if self.ndims == 1:

tests/test_import_export_data.py

@@ -126,7 +126,7 @@ def test_import_data_array(sample_array):
     data_filename = os.path.join(os.path.dirname(__file__), "data", "matrix.tns")
     X = ttb.import_data(data_filename)
 
-    assert (M == X).all()
+    assert np.array_equal(M, X)
 
 
 @pytest.mark.indevelopment
@@ -230,14 +230,14 @@ def test_export_data_array(sample_array):
     ttb.export_data(M, data_filename)
 
     X = ttb.import_data(data_filename)
-    assert (M == X).all()
+    assert np.array_equal(M, X)
     os.unlink(data_filename)
 
     data_filename = os.path.join(os.path.dirname(__file__), "data", "matrix_int.out")
     ttb.export_data(M, data_filename, fmt_data="%d")
 
     X = ttb.import_data(data_filename)
-    assert (M == X).all()
+    assert np.array_equal(M, X)
     os.unlink(data_filename)
 
 

tests/test_khatrirao.py

@@ -24,9 +24,9 @@ def test_khatrirao():
             [64, 125, 216],
         ]
     )
-    assert (ttb.khatrirao(*[A, A, A]) == answer).all()
-    assert (ttb.khatrirao(*[A, A, A], reverse=True) == answer).all()
-    assert (ttb.khatrirao(A, A, A) == answer).all()
+    assert np.array_equal(ttb.khatrirao(*[A, A, A]), answer)
+    assert np.array_equal(ttb.khatrirao(*[A, A, A], reverse=True), answer)
+    assert np.array_equal(ttb.khatrirao(A, A, A), answer)
 
     # Test case where inputs are column vectors
     a_1 = np.array([[1], [1], [1], [1]])
@@ -40,8 +40,8 @@ def test_khatrirao():
             a_2[3, 0] * np.ones((16, 1)),
         )
     )
-    assert (ttb.khatrirao(*[a_2, a_1, a_1]) == result).all()
-    assert (ttb.khatrirao(a_2, a_1, a_1) == result).all()
+    assert np.array_equal(ttb.khatrirao(*[a_2, a_1, a_1]), result)
+    assert np.array_equal(ttb.khatrirao(a_2, a_1, a_1), result)
 
     with pytest.raises(AssertionError) as excinfo:
         ttb.khatrirao(a_2, a_1, np.ones((2, 2, 2)))