1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
|
from typing import Tuple
import _numpy_minmax
import numpy as np
from numpy.typing import NDArray
try:
from ._version import version as __version__
except ImportError:
__version__ = "unknown"
def minmax(a: NDArray) -> Tuple:
if 0 in a.shape:
raise ValueError("Cannot find min/max value in empty array")
if a.dtype == np.dtype("float32"):
if a.flags["C_CONTIGUOUS"] or a.flags["F_CONTIGUOUS"]:
result = _numpy_minmax.lib.minmax_contiguous_float32(
_numpy_minmax.ffi.cast("float *", a.ctypes.data), a.size
)
return np.float32(result.min_val), np.float32(result.max_val)
if a.ndim == 1:
result = _numpy_minmax.lib.minmax_1d_strided_float32(
_numpy_minmax.ffi.cast("float *", a.ctypes.data), a.size, a.strides[0]
)
return np.float32(result.min_val), np.float32(result.max_val)
# TODO: Find multi-dim arrays that can be simplified to a single stride
elif a.dtype == np.dtype("int16") and (
a.flags["C_CONTIGUOUS"] or a.flags["F_CONTIGUOUS"]
):
result = _numpy_minmax.lib.minmax_contiguous_int16(
_numpy_minmax.ffi.cast("int16_t *", a.ctypes.data), a.size
)
return np.int16(result.min_val), np.int16(result.max_val)
return np.amin(a), np.amax(a)
|
