Add new measurements

Author: FrancescAlted

bench/ndarray/jit-reduc-float64-plot-semilogx.py

@@ -39,17 +39,17 @@
     create_BLOSCLZ_l7_disk = [ 0.0701, 0.2656, 1.0553, 4.0486, 9.2255, 12.2674, 16.4618, 20.1527, 25.3657, 33.7537, 37.3551, 43.0586, 48.4968, 53.9183, 62.9415, 71.7656, 80.5597, 85.5704, 97.0770, 109.7463, 119.2675 ]
     compute_BLOSCLZ_l7_disk = [ 0.0019, 0.0213, 0.0788, 0.3002, 0.7252, 0.9276, 1.2053, 1.4999, 1.9109, 3.4081, 2.8205, 3.3593, 3.6086, 4.2295, 4.9548, 5.6996, 6.0085, 7.0802, 7.4786, 8.4466, 9.4861 ]
 
-    create_LZ4_l1 = [ 0.0304, 0.2582, 1.0298, 3.9502, 8.8945, 11.9267, 16.3965, 20.2368, 24.6837, 29.3425, 36.2631, 42.1709, 48.0605, 52.3962, 61.5175, 68.6328, 80.1160, 85.4322, 97.1122, 106.9973, 114.8584, 219.8679, 372.3182, 650.2087, 876.6964, 1535.3019, 1717.6310 ]
-    compute_LZ4_l1 = [ 0.0018, 0.0210, 0.0756, 0.3003, 0.6609, 0.8886, 1.1285, 1.4453, 1.7959, 2.1889, 2.6978, 3.1586, 3.4286, 3.9929, 4.4590, 5.3601, 5.6702, 6.4690, 6.9764, 7.8714, 8.6404, 15.7214, 29.4130, 46.5909, 87.1930, 164.6234, 258.9626 ]
+    create_LZ4_l1 = [ 0.0304, 0.2582, 1.0298, 3.9502, 8.8945, 11.9267, 16.3965, 20.2368, 24.6837, 29.3425, 36.2631, 42.1709, 48.0605, 52.3962, 61.5175, 68.6328, 80.1160, 85.4322, 97.1122, 106.9973, 114.8584, 219.8679, 372.3182, 650.2087, 876.6964, 1535.3019, 1717.6310, 2605.6513, 3490.7571, 4253.5521, 4192.6208, 6181.3742, 6793.9787 ]
+    compute_LZ4_l1 = [ 0.0018, 0.0210, 0.0756, 0.3003, 0.6609, 0.8886, 1.1285, 1.4453, 1.7959, 2.1889, 2.6978, 3.1586, 3.4286, 3.9929, 4.4590, 5.3601, 5.6702, 6.4690, 6.9764, 7.8714, 8.6404, 15.7214, 29.4130, 46.5909, 87.1930, 164.6234, 258.9626, 256.4864, 378.0102, 476.1793, 585.9910, 734.2687, 853.7598 ]
 
-    create_LZ4_l1_disk = [ 1.7980, 0.2617, 1.0480, 4.0809, 9.0720, 13.8294, 16.7269, 20.5108, 24.9465, 30.0428, 37.1903, 42.8075, 48.7775, 52.9890, 63.4071, 70.1766, 81.9747, 88.1830, 97.7921, 111.0611, 119.7673, 214.8363, 370.7900, 600.6060, 872.7770, 1314.0561, 1581.3989 ]
-    compute_LZ4_l1_disk = [ 0.0019, 0.0214, 0.0795, 0.3060, 0.6985, 0.9195, 1.1766, 1.5213, 1.8845, 2.2972, 2.8044, 3.2587, 3.5898, 4.1524, 4.6293, 5.5485, 5.8715, 6.7386, 7.3019, 8.2307, 9.0145, 16.1475, 30.1677, 59.1110, 81.9494, 112.0279, 169.0670 ]
+    create_LZ4_l1_disk = [ 1.7980, 0.2617, 1.0480, 4.0809, 9.0720, 13.8294, 16.7269, 20.5108, 24.9465, 30.0428, 37.1903, 42.8075, 48.7775, 52.9890, 63.4071, 70.1766, 81.9747, 88.1830, 97.7921, 111.0611, 119.7673, 214.8363, 370.7900, 600.6060, 872.7770, 1314.0561, 1581.3989, 1898.3007, 2910.3205, 3476.1479, 4753.6958, 5590.7596, 6627.1739 ]
+    compute_LZ4_l1_disk = [ 0.0019, 0.0214, 0.0795, 0.3060, 0.6985, 0.9195, 1.1766, 1.5213, 1.8845, 2.2972, 2.8044, 3.2587, 3.5898, 4.1524, 4.6293, 5.5485, 5.8715, 6.7386, 7.3019, 8.2307, 9.0145, 16.1475, 30.1677, 59.1110, 81.9494, 112.0279, 169.0670, 173.9750, 248.5645, 332.5040, 354.8242, 448.8191, 493.8022 ]
 
-    create_ZSTD_l1 = [ 0.0302, 0.2704, 1.0703, 4.1243, 9.2185, 12.5026, 17.0585, 20.8708, 25.5844, 31.0571, 37.7114, 42.8297, 50.2696, 54.5773, 63.6311, 73.0370, 84.0092, 89.0686, 100.3300, 108.8173, 119.1154 ]
-    compute_ZSTD_l1 = [ 0.0021, 0.0296, 0.1045, 0.3979, 0.8787, 1.3064, 1.7404, 2.1938, 2.6780, 3.3929, 3.8601, 4.3665, 5.0127, 5.7346, 6.1056, 7.9448, 8.2872, 9.4659, 9.2376, 10.4273, 11.6572 ]
+    create_ZSTD_l1 = [ 0.0302, 0.2704, 1.0703, 4.1243, 9.2185, 12.5026, 17.0585, 20.8708, 25.5844, 31.0571, 37.7114, 42.8297, 50.2696, 54.5773, 63.6311, 73.0370, 84.0092, 89.0686, 100.3300, 108.8173, 119.1154, 265.8825, 493.3042, 851.1048, 1165.6934, 1589.0762, 2055.2161, 2481.3166, 3501.0184, 4258.2440, 4151.9682, 6119.5858, 6518.2127 ]
+    compute_ZSTD_l1 = [ 0.0021, 0.0296, 0.1045, 0.3979, 0.8787, 1.3064, 1.7404, 2.1938, 2.6780, 3.3929, 3.8601, 4.3665, 5.0127, 5.7346, 6.1056, 7.9448, 8.2872, 9.4659, 9.2376, 10.4273, 11.6572, 22.0410, 36.7011, 65.2484, 84.9773, 123.1597, 147.6101, 274.7479, 384.7447, 442.0842, 512.2530, 641.9793, 702.5878 ]
 
-    create_ZSTD_l1_disk = [ 0.6564, 0.2825, 1.0826, 4.1968, 9.5022, 13.4840, 17.5387, 21.5807, 26.0052, 31.3524, 38.5889, 44.1105, 49.8849, 55.5297, 64.6479, 72.7471, 84.6595, 90.4970, 99.9710, 111.6817, 120.8941 ]
-    compute_ZSTD_l1_disk = [ 0.0022, 0.0300, 0.1066, 0.4099, 0.8974, 1.3218, 1.7679, 2.2154, 2.7007, 3.4267, 3.9255, 4.4597, 5.1155, 5.8251, 6.2064, 8.0141, 8.4316, 9.3195, 9.4570, 10.7034, 11.9192 ]
+    create_ZSTD_l1_disk = [ 0.6564, 0.2825, 1.0826, 4.1968, 9.5022, 13.4840, 17.5387, 21.5807, 26.0052, 31.3524, 38.5889, 44.1105, 49.8849, 55.5297, 64.6479, 72.7471, 84.6595, 90.4970, 99.9710, 111.6817, 120.8941, 234.9739, 391.9157, 648.5382, 920.2396, 1367.7080, 1647.1145, 2440.9581, 3028.6825, 3518.1483, 4601.6684, 5660.8254, 6723.2414 ]
+    compute_ZSTD_l1_disk = [ 0.0022, 0.0300, 0.1066, 0.4099, 0.8974, 1.3218, 1.7679, 2.2154, 2.7007, 3.4267, 3.9255, 4.4597, 5.1155, 5.8251, 6.2064, 8.0141, 8.4316, 9.3195, 9.4570, 10.7034, 11.9192, 22.1895, 36.6542, 66.7209, 89.2111, 126.3853, 155.7241, 203.4894, 288.3248, 352.8067, 383.0908, 478.0074, 545.8722 ]
 
     create_numpy = [ 0.0020, 0.0527, 0.2292, 0.9412, 2.1043, 2.8286, 3.7046, 4.7217, 5.8308, 7.0491 ]
     compute_numpy = [ 0.0179, 0.2495, 0.9840, 3.9263, 8.8450, 12.0259, 16.3507, 40.1672, 155.1292, 302.5115 ]

doc/reference/linear_algebra.rst

@@ -12,3 +12,4 @@ The next functions can be used for computing linear algebra operations with :ref
    :nosignatures:
 
     matmul
+    transpose

src/blosc2/__init__.py

@@ -248,6 +248,7 @@ class Tuner(Enum):
     full,
     save,
     matmul,
+    transpose,
 )
 
 from .c2array import c2context, C2Array, URLPath

src/blosc2/ndarray.py

@@ -3755,6 +3755,49 @@ def matmul(x1: NDArray, x2: NDArray, **kwargs: Any) -> NDArray:
     return result.squeeze()
 
 
+def transpose(x, **kwargs: Any) -> NDArray:
+    """
+    Returns a Blosc2 NDArray with axes transposed.
+
+    Parameters
+    ----------
+    x: `NDArray`
+        The input matrix.
+    kwargs: Any, optional
+        Keyword arguments that are supported by the :func:`empty` constructor.
+
+    Returns
+    -------
+    out: :ref:`NDArray`
+        The Blosc2 NDArray with axes transposed.
+
+    References
+    ----------
+    `numpy.transpose <https://numpy.org/doc/2.2/reference/generated/numpy.transpose.html>`_
+    """
+
+    # Validate arguments are not scalars
+    if np.isscalar(x) or x.ndim < 2:
+        return x
+
+    # Validate arguments are dimension 2
+    if x.ndim > 2:
+        raise ValueError("Transposing arrays with dimension greater than 2 is not supported yet.")
+
+    n, m = x.shape
+    p, q = x.chunks
+    result = blosc2.zeros((m, n), dtype=np.result_type(x), **kwargs)
+
+    for row in range(0, n, p):
+        row_end = (row + p) if (row + p) < n else n
+        for col in range(0, m, q):
+            col_end = (col + q) if (col + q) < m else m
+            aux = x[row:row_end, col:col_end]
+            result[col:col_end, row:row_end] = np.transpose(aux).copy()
+
+    return result
+
+
 # Class for dealing with fields in an NDArray
 # This will allow to access fields by name in the dtype of the NDArray
 class NDField(Operand):

tests/ndarray/test_matmul.py

@@ -21,11 +21,50 @@
 )
 @pytest.mark.parametrize(
     "dtype",
-    {np.float32, np.float64, np.complex64, np.complex128},
+    {np.float32, np.float64},
 )
 def test_matmul(ashape, achunks, ablocks, bshape, bchunks, bblocks, dtype):
-    a = blosc2.linspace(0, 10, dtype=dtype, shape=ashape, chunks=achunks, blocks=ablocks)
-    b = blosc2.linspace(0, 10, dtype=dtype, shape=bshape, chunks=bchunks, blocks=bblocks)
+    a = blosc2.linspace(0, 1, dtype=dtype, shape=ashape, chunks=achunks, blocks=ablocks)
+    b = blosc2.linspace(0, 1, dtype=dtype, shape=bshape, chunks=bchunks, blocks=bblocks)
+    c = blosc2.matmul(a, b)
+
+    na = a[:]
+    nb = b[:]
+    nc = np.matmul(na, nb)
+
+    np.testing.assert_allclose(c, nc, rtol=1e-6)
+
+
+@pytest.mark.parametrize(
+    ("ashape", "achunks", "ablocks"),
+    {
+        ((12, 10), (7, 5), (3, 3)),
+        ((10,), (9,), (7,)),
+    },
+)
+@pytest.mark.parametrize(
+    ("bshape", "bchunks", "bblocks"),
+    {
+        ((10,), (4,), (2,)),
+        ((10, 5), (3, 4), (1, 3)),
+        ((10, 12), (2, 4), (1, 2)),
+    },
+)
+@pytest.mark.parametrize(
+    "dtype",
+    {np.complex64, np.complex128},
+)
+def test_complex(ashape, achunks, ablocks, bshape, bchunks, bblocks, dtype):
+    real_part = blosc2.linspace(0, 1, shape=ashape, chunks=achunks, blocks=ablocks, dtype=dtype)
+    imag_part = blosc2.linspace(0, 1, shape=ashape, chunks=achunks, blocks=ablocks, dtype=dtype)
+    complex_matrix_a = real_part + 1j * imag_part
+    a = blosc2.asarray(complex_matrix_a)
+
+    real_part = blosc2.linspace(1, 2, shape=bshape, chunks=bchunks, blocks=bblocks, dtype=dtype)
+    imag_part = blosc2.linspace(1, 2, shape=bshape, chunks=bchunks, blocks=bblocks, dtype=dtype)
+    complex_matrix_b = real_part + 1j * imag_part
+    b = blosc2.asarray(complex_matrix_b)
+
     c = blosc2.matmul(a, b)
 
     na = a[:]

tests/ndarray/test_transpose.py

@@ -0,0 +1,108 @@
+import numpy as np
+import pytest
+
+import blosc2
+
+
+@pytest.mark.parametrize(
+    ("shape", "chunks", "blocks"),
+    {
+        ((3, 3), (2, 2), (1, 1)),
+        ((12, 11), (7, 5), (6, 2)),
+        ((1, 5), (1, 4), (1, 3)),
+        ((201, 1903), None, None),
+        ((10,), (5,), None),
+    },
+)
+@pytest.mark.parametrize(
+    "dtype",
+    {np.int32, np.int64, np.float32, np.float64},
+)
+def test_transpose(shape, chunks, blocks, dtype):
+    a = blosc2.linspace(0, 1, shape=shape, chunks=chunks, blocks=blocks, dtype=dtype)
+    at = blosc2.transpose(a)
+
+    na = a[:]
+    nat = np.transpose(na)
+
+    np.testing.assert_allclose(at, nat)
+
+
+@pytest.mark.parametrize(
+    ("shape", "chunks", "blocks"),
+    {
+        ((3, 3), (2, 2), (1, 1)),
+        ((12, 11), (7, 5), (6, 2)),
+        ((1, 5), (1, 4), (1, 3)),
+        ((201, 1903), None, None),
+        ((10,), (5,), None),
+    },
+)
+@pytest.mark.parametrize(
+    "dtype",
+    {np.complex64, np.complex128},
+)
+def test_complex(shape, chunks, blocks, dtype):
+    real_part = blosc2.linspace(0, 1, shape=shape, chunks=chunks, blocks=blocks, dtype=dtype)
+    imag_part = blosc2.linspace(0, 1, shape=shape, chunks=chunks, blocks=blocks, dtype=dtype)
+    complex_matrix = real_part + 1j * imag_part
+
+    a = blosc2.asarray(complex_matrix)
+    at = blosc2.transpose(a)
+
+    na = a[:]
+    nat = np.transpose(na)
+
+    np.testing.assert_allclose(at, nat)
+
+
+@pytest.mark.parametrize(
+    "scalar",
+    {
+        1,  # int
+        5.1,  # float
+        1 + 2j,  # complex
+        np.int8(2),  # NumPy int8
+        np.int16(3),  # NumPy int16
+        np.int32(4),  # NumPy int32
+        np.int64(5),  # NumPy int64
+        np.float32(5.2),  # NumPy float32
+        np.float64(5.3),  # NumPy float64
+        np.complex64(0 + 3j),  # NumPy complex64
+        np.complex128(2 - 4j),  # NumPy complex128
+    },
+)
+def test_scalars(scalar):
+    at = blosc2.transpose(scalar)
+    nat = np.transpose(scalar)
+
+    np.testing.assert_allclose(at, nat)
+
+
+@pytest.mark.parametrize(
+    "shape",
+    [
+        (3, 3, 3),
+        (12, 10, 10),
+        (10, 10, 10, 11),
+        (5, 4, 3, 2, 1, 0),
+    ],
+)
+def test_dims(shape):
+    a = blosc2.linspace(0, 1, shape=shape)
+
+    with pytest.raises(ValueError):
+        blosc2.transpose(a)
+
+
+def test_disk():
+    a = blosc2.linspace(0, 1, shape=(3, 4), urlpath="a_test.b2nd", mode="w")
+    c = blosc2.transpose(a, urlpath="c_test.b2nd", mode="w")
+
+    na = a[:]
+    nc = np.transpose(na)
+
+    np.testing.assert_allclose(c, nc, rtol=1e-6)
+
+    blosc2.remove_urlpath("a_test.b2nd")
+    blosc2.remove_urlpath("c_test.b2nd")