【Hackathon 5th No.37】为 Paddle 新增 householder_product API -part

python/paddle/linalg.py

@@ -12,33 +12,32 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from .tensor import inverse as inv
-from .tensor.linalg import (
-    cholesky,
-    cholesky_solve,
-    cond,
-    corrcoef,
-    cov,
-    det,
-    eig,
-    eigh,
-    eigvals,
-    eigvalsh,
-    lstsq,
-    lu,
-    lu_unpack,
-    matrix_power,
-    matrix_rank,
-    multi_dot,
-    norm,
-    pca_lowrank,
-    pinv,
-    qr,
-    slogdet,
-    solve,
-    svd,
-    triangular_solve,
-)
+from .tensor import inverse as inv  # noqa: F401
+from .tensor.linalg import cholesky  # noqa: F401
+from .tensor.linalg import cholesky_solve  # noqa: F401
+from .tensor.linalg import cond  # noqa: F401
+from .tensor.linalg import corrcoef  # noqa: F401
+from .tensor.linalg import cov  # noqa: F401
+from .tensor.linalg import det  # noqa: F401
+from .tensor.linalg import eig  # noqa: F401
+from .tensor.linalg import eigh  # noqa: F401
+from .tensor.linalg import eigvals  # noqa: F401
+from .tensor.linalg import eigvalsh  # noqa: F401
+from .tensor.linalg import householder_product  # noqa: F401
+from .tensor.linalg import lu  # noqa: F401
+from .tensor.linalg import lu_unpack  # noqa: F401
+from .tensor.linalg import matrix_power  # noqa: F401
+from .tensor.linalg import matrix_rank  # noqa: F401
+from .tensor.linalg import multi_dot  # noqa: F401
+from .tensor.linalg import norm  # noqa: F401
+from .tensor.linalg import pca_lowrank  # noqa: F401
+from .tensor.linalg import pinv  # noqa: F401
+from .tensor.linalg import qr  # noqa: F401
+from .tensor.linalg import slogdet  # noqa: F401
+from .tensor.linalg import solve  # noqa: F401
+from .tensor.linalg import svd  # noqa: F401
+from .tensor.linalg import triangular_solve  # noqa: F401
+from .tensor.linalg import lstsq
 
 __all__ = [
     'cholesky',
@@ -53,6 +52,7 @@
     'matrix_rank',
     'svd',
     'qr',
+    'householder_product',
     'pca_lowrank',
     'lu',
     'lu_unpack',

python/paddle/tensor/__init__.py

@@ -68,6 +68,7 @@
 from .linalg import eig  # noqa: F401
 from .linalg import matrix_power  # noqa: F401
 from .linalg import qr  # noqa: F401
+from .linalg import householder_product  # noqa: F401
 from .linalg import eigvals  # noqa: F401
 from .linalg import multi_dot  # noqa: F401
 from .linalg import svd  # noqa: F401
@@ -413,6 +414,7 @@
     'mv',
     'matrix_power',
     'qr',
+    'householder_product',
     'pca_lowrank',
     'eigvals',
     'eigvalsh',

python/paddle/tensor/linalg.py

@@ -3724,3 +3724,120 @@ def cdist(
     return paddle.linalg.norm(
         x[..., None, :] - y[..., None, :, :], p=p, axis=-1
     )
+
+
+def householder_product(A, tau, name=None):
+    r"""
+
+    Computes the first n columns of a product of Householder matrices.
+
+    This function can get the vector :math:`\omega_{i}` from matrix `A`(m x n), the :math:`i-1` elements are zeros, and the i-th is `1`, the rest of the elements are from i-th column of `A`.
+    And with the vector `tau` can calculate the first n columns of a product of Householder matrices.
+
+    :math:`H_i = I_m - \tau_i \omega_i \omega_i^H`
+
+    Args:
+        A (Tensor): A tensor with shape (*, m, n) where * is zero or more batch dimensions.
+        tau (Tensor): A tensor with shape (*, k) where * is zero or more batch dimensions.
+        name (str, optional): For details, please refer to :ref:`api_guide_Name`. Generally, no setting is required. Default: None.
+
+    Returns:
+        Tensor, the dtype is same as input tensor, the Q in QR decomposition.
+
+        :math:`out = Q = H_1H_2H_3...H_k`
+
+    Examples:
+        .. code-block:: python
+
+            >>> import paddle
+            >>> A = paddle.to_tensor([[-1.1280,  0.9012, -0.0190],
+            ...         [ 0.3699,  2.2133, -1.4792],
+            ...         [ 0.0308,  0.3361, -3.1761],
+            ...         [-0.0726,  0.8245, -0.3812]])
+            >>> tau = paddle.to_tensor([1.7497, 1.1156, 1.7462])
+            >>> Q = paddle.linalg.householder_product(A, tau)
+            >>> Q
+            Tensor(shape=[4, 3], dtype=float32, place=Place(gpu:0), stop_gradient=True,
+                   [[-0.74969995, -0.02181768,  0.31115776],
+                    [-0.64721400, -0.12367040, -0.21738708],
+                    [-0.05389076, -0.37562513, -0.84836429],
+                    [ 0.12702821, -0.91822827,  0.36892807]])
+    """
+
+    check_dtype(
+        A.dtype,
+        'x',
+        [
+            'float32',
+            'float64',
+        ],
+        'householder_product',
+    )
+    check_dtype(
+        tau.dtype,
+        'tau',
+        [
+            'float32',
+            'float64',
+        ],
+        'householder_product',
+    )
+    assert (
+        A.dtype == tau.dtype
+    ), "The input A must have the same dtype with input tau.\n"
+    assert (
+        len(A.shape) >= 2
+        and len(tau.shape) >= 1
+        and len(A.shape) == len(tau.shape) + 1
+    ), (
+        "The input A must have more than 2 dimensions, and input tau must have more than 1 dimension,"
+        "and the dimension of A is 1 larger than the dimension of tau\n"
+    )
+    assert (
+        A.shape[-2] >= A.shape[-1]
+    ), "The rows of input A must be greater than or equal to the columns of input A.\n"
+    for idx, _ in enumerate(A.shape[:-2]):
+        assert (
+            A.shape[idx] == tau.shape[idx]
+        ), "The input A must have the same batch dimensions with input tau.\n"
+
+    def _householder_product(A, tau):
+        m, n = A.shape[-2:]
+        Q = paddle.eye(m)
+        for i in range(min(m, n)):
+            normx = paddle.norm(A[i:, i])
+            sign = 1 if A[i, i] < 0 else -1
+            w = A[i:, i]
+            if in_dynamic_mode():
+                w[0] = 1
+            else:
+                w = paddle.static.setitem(w, 0, 1)
+            w = w.reshape([-1, 1])
+            if in_dynamic_mode():
+                Q[:, i:] = Q[:, i:] - (Q[:, i:] @ w @ w.T * tau[i])
+            else:
+                Q = paddle.static.setitem(
+                    Q,
+                    (slice(None), slice(i, None)),
+                    Q[:, i:] - (Q[:, i:] @ w @ w.T * tau[i]),
+                )
+        return Q[:, :n]
+
+    if len(A.shape) == 2:
+        return _householder_product(A, tau)
+    m, n = A.shape[-2:]
+    org_A_shape = A.shape
+    org_tau_shape = tau.shape
+    A = A.reshape((-1, org_A_shape[-2], org_A_shape[-1]))
+    tau = tau.reshape((-1, org_tau_shape[-1]))
+    n_batch = A.shape[0]
+    out = paddle.zeros([n_batch, m, n])
+    for i in range(n_batch):
+        if in_dynamic_mode():
+            out[i] = _householder_product(A[i], tau[i])
+        else:
+            out = paddle.static.setitem(
+                out, i, _householder_product(A[i], tau[i])
+            )
+    out = out.reshape(org_A_shape)
+    return out