How to use the tensorly.backend.ndim function in tensorly

Returns
        -------
        ndarray
            `mode`-mode product of `tensor` by `matrix_or_vector`
            * of shape :math:`(i_1, ..., i_{k-1}, J, i_{k+1}, ..., i_N)` if matrix_or_vector is a matrix
            * of shape :math:`(i_1, ..., i_{k-1}, i_{k+1}, ..., i_N)` if matrix_or_vector is a vector

        See also
        --------
        multi_mode_dot : chaining several mode_dot in one call
        """
        # the mode along which to fold might decrease if we take product with a vector
        fold_mode = mode
        new_shape = list(tensor.shape)

        if T.ndim(matrix_or_vector) == 2:  # Tensor times matrix
            # Test for the validity of the operation
            if matrix_or_vector.shape[1] != tensor.shape[mode]:
                raise ValueError(
                    'shapes {0} and {1} not aligned in mode-{2} multiplication: {3} (mode {2}) != {4} (dim 1 of matrix)'.format(
                        tensor.shape, matrix_or_vector.shape, mode, tensor.shape[mode], matrix_or_vector.shape[1]
                    ))
            new_shape[mode] = matrix_or_vector.shape[0]
            vec = False

        elif T.ndim(matrix_or_vector) == 1:  # Tensor times vector
            if matrix_or_vector.shape[0] != tensor.shape[mode]:
                raise ValueError(
                    'shapes {0} and {1} not aligned for mode-{2} multiplication: {3} (mode {2}) != {4} (vector size)'.format(
                        tensor.shape, matrix_or_vector.shape, mode, tensor.shape[mode], matrix_or_vector.shape[0]
                    ))
            if len(new_shape) > 1:

Returns
        -------
        KruskalTensor = (core, factors)
            `mode`-mode product of `tensor` by `matrix_or_vector`
            * of shape :math:`(i_1, ..., i_{k-1}, J, i_{k+1}, ..., i_N)` if matrix_or_vector is a matrix
            * of shape :math:`(i_1, ..., i_{k-1}, i_{k+1}, ..., i_N)` if matrix_or_vector is a vector

        See also
        --------
        kruskal_multi_mode_dot : chaining several mode_dot in one call
        """
        shape, _ = _validate_kruskal_tensor(kruskal_tensor)
        weights, factors = kruskal_tensor
        contract = False
        
        if T.ndim(matrix_or_vector) == 2:  # Tensor times matrix
            # Test for the validity of the operation
            if matrix_or_vector.shape[1] != shape[mode]:
                raise ValueError(
                    'shapes {0} and {1} not aligned in mode-{2} multiplication: {3} (mode {2}) != {4} (dim 1 of matrix)'.format(
                        shape, matrix_or_vector.shape, mode, shape[mode], matrix_or_vector.shape[1]
                    ))

        elif T.ndim(matrix_or_vector) == 1:  # Tensor times vector
            if matrix_or_vector.shape[0] != shape[mode]:
                raise ValueError(
                    'shapes {0} and {1} not aligned for mode-{2} multiplication: {3} (mode {2}) != {4} (vector size)'.format(
                        shape, matrix_or_vector.shape, mode, shape[mode], matrix_or_vector.shape[0]
                    ))
            if not keep_dim:
                contract = True # Contract over that mode
        else:

override_module_dispatch(__name__, _get_backend_method, _get_backend_dir)

def dispatch_sparse(func):
    @functools.wraps(func, assigned=('__name__', '__qualname__',
                                     '__doc__', '__annotations__'))
    def inner(*args, **kwargs):
        with sparse_context():
            return func(*args, **kwargs)

    return inner

tensor = dispatch_sparse(backend.tensor)
is_tensor = dispatch_sparse(backend.is_tensor)
context = dispatch_sparse(backend.context)
shape = dispatch_sparse(backend.shape)
ndim = dispatch_sparse(backend.ndim)
to_numpy = dispatch_sparse(backend.to_numpy)
copy = dispatch_sparse(backend.copy)
concatenate = dispatch_sparse(backend.concatenate)
reshape = dispatch_sparse(backend.reshape)
moveaxis = dispatch_sparse(backend.moveaxis)
transpose = dispatch_sparse(backend.transpose)
arange = dispatch_sparse(backend.arange)
ones = dispatch_sparse(backend.ones)
zeros = dispatch_sparse(backend.zeros)
zeros_like = dispatch_sparse(backend.zeros_like)
eye = dispatch_sparse(backend.eye,)
clip = dispatch_sparse(backend.clip)
where = dispatch_sparse(backend.where)
max = dispatch_sparse(backend.max)
min = dispatch_sparse(backend.min)
all = dispatch_sparse(backend.all)

elif T.ndim(matrix_or_vector) == 1:  # Tensor times vector
            if matrix_or_vector.shape[0] != tensor.shape[mode]:
                raise ValueError(
                    'shapes {0} and {1} not aligned for mode-{2} multiplication: {3} (mode {2}) != {4} (vector size)'.format(
                        tensor.shape, matrix_or_vector.shape, mode, tensor.shape[mode], matrix_or_vector.shape[0]
                    ))
            if len(new_shape) > 1:
                new_shape.pop(mode)
            else:
                new_shape = [1]
            vec = True

        else:
            raise ValueError('Can only take n_mode_product with a vector or a matrix.'
                             'Provided array of dimension {} not in [1, 2].'.format(T.ndim(matrix_or_vector)))

        res = T.dot(matrix_or_vector, unfold(tensor, mode))

        if vec: # We contracted with a vector, leading to a vector
            return vec_to_tensor(res, shape=new_shape)
        else: # tensor times vec: refold the unfolding
            return fold(res, fold_mode, new_shape)

"""
        # the mode along which to fold might decrease if we take product with a vector
        fold_mode = mode
        new_shape = list(tensor.shape)

        if T.ndim(matrix_or_vector) == 2:  # Tensor times matrix
            # Test for the validity of the operation
            if matrix_or_vector.shape[1] != tensor.shape[mode]:
                raise ValueError(
                    'shapes {0} and {1} not aligned in mode-{2} multiplication: {3} (mode {2}) != {4} (dim 1 of matrix)'.format(
                        tensor.shape, matrix_or_vector.shape, mode, tensor.shape[mode], matrix_or_vector.shape[1]
                    ))
            new_shape[mode] = matrix_or_vector.shape[0]
            vec = False

        elif T.ndim(matrix_or_vector) == 1:  # Tensor times vector
            if matrix_or_vector.shape[0] != tensor.shape[mode]:
                raise ValueError(
                    'shapes {0} and {1} not aligned for mode-{2} multiplication: {3} (mode {2}) != {4} (vector size)'.format(
                        tensor.shape, matrix_or_vector.shape, mode, tensor.shape[mode], matrix_or_vector.shape[0]
                    ))
            if len(new_shape) > 1:
                new_shape.pop(mode)
            else:
                new_shape = [1]
            vec = True

        else:
            raise ValueError('Can only take n_mode_product with a vector or a matrix.'
                             'Provided array of dimension {} not in [1, 2].'.format(T.ndim(matrix_or_vector)))

        res = T.dot(matrix_or_vector, unfold(tensor, mode))

Returns
        -------
        TuckerTensor = (core, factors)
            `mode`-mode product of `tensor` by `matrix_or_vector`
            * of shape :math:`(i_1, ..., i_{k-1}, J, i_{k+1}, ..., i_N)` if matrix_or_vector is a matrix
            * of shape :math:`(i_1, ..., i_{k-1}, i_{k+1}, ..., i_N)` if matrix_or_vector is a vector

        See also
        --------
        tucker_multi_mode_dot : chaining several mode_dot in one call
        """
        shape, rank = _validate_tucker_tensor(tucker_tensor)
        core, factors = tucker_tensor
        contract = False
        
        if tl.ndim(matrix_or_vector) == 2:  # Tensor times matrix
            # Test for the validity of the operation
            if matrix_or_vector.shape[1] != shape[mode]:
                raise ValueError(
                    'shapes {0} and {1} not aligned in mode-{2} multiplication: {3} (mode {2}) != {4} (dim 1 of matrix)'.format(
                        shape, matrix_or_vector.shape, mode, shape[mode], matrix_or_vector.shape[1]
                    ))

        elif tl.ndim(matrix_or_vector) == 1:  # Tensor times vector
            if matrix_or_vector.shape[0] != shape[mode]:
                raise ValueError(
                    'shapes {0} and {1} not aligned for mode-{2} multiplication: {3} (mode {2}) != {4} (vector size)'.format(
                        shape, matrix_or_vector.shape, mode, shape[mode], matrix_or_vector.shape[0]
                    ))
            if not keep_dim:
                contract = True # Contract over that mode
        else:

# Initialise the decompositions
    D = T.zeros_like(X, **T.context(X))  # low rank part
    E = T.zeros_like(X, **T.context(X))  # sparse part
    L_x = T.zeros_like(X, **T.context(X))  # Lagrangian variables for the (X - D - E - L_x/mu) term
    J = [T.zeros_like(X, **T.context(X)) for _ in range(T.ndim(X))] # Low-rank modes of X
    L = [T.zeros_like(X, **T.context(X)) for _ in range(T.ndim(X))] # Lagrangian or J

    # Norm of the reconstructions at each iteration
    rec_X = []
    rec_D = []

    mu = mu_init

    for iteration in range(n_iter_max):

        for i in range(T.ndim(X)):
            J[i] = fold(svd_thresholding(unfold(D, i) + unfold(L[i], i)/mu, reg_J/mu), i, X.shape)

        D = L_x/mu + X - E
        for i in range(T.ndim(X)):
            D += J[i] - L[i]/mu
        D /= (T.ndim(X) + 1)

        E = soft_thresholding(X - D + L_x/mu, mask*reg_E/mu)

        # Update the lagrangian multipliers
        for i in range(T.ndim(X)):
            L[i] += mu * (D - J[i])

        L_x += mu*(X - D - E)

        mu = min(mu*learning_rate, mu_max)

----------
        X : ndarray of shape (n_samples, N1, ..., NS)
            tensor data
        y : array of shape (n_samples)
            labels associated with each sample

        Returns
        -------
        self
        """
        rng = check_random_state(self.random_state)

        # Initialise randomly the weights
        G = T.tensor(rng.randn(*self.weight_ranks), **T.context(X))
        W = []
        for i in range(1, T.ndim(X)):  # First dimension of X = number of samples
            W.append(T.tensor(rng.randn(X.shape[i], G.shape[i - 1]), **T.context(X)))

        # Norm of the weight tensor at each iteration
        norm_W = []

        for iteration in range(self.n_iter_max):

            # Optimise modes of W
            for i in range(len(W)):
                phi = partial_tensor_to_vec(
                            T.dot(partial_unfold(X, i),
                                  T.dot(kronecker(W, skip_matrix=i),
                                          T.transpose(unfold(G, i)))))
                # Regress phi on y: we could call a package here, e.g. scikit-learn
                inv_term = T.dot(T.transpose(phi), phi) +\
                     self.reg_W * T.tensor(np.eye(phi.shape[1]), **T.context(X))