Class NDLog1pExpm1Math
Scalar kernels backing NumSharp's float32 / float64 np.log1p and np.expm1, and the log1p inside
np.logaddexp / np.logaddexp2 (NDLogAddExpMath): models of the MSVC UCRT routines that
NumPy 2.4.2's win-amd64 wheel calls (npy_log1p / npy_expm1 forward to the C library; the AVX-512 SVML
loops that would replace them are not built into that wheel).
log1p / log1pf are BIT-EXACT with NumPy. Probed against np.log1p (parity QA-22, 2026-11-05): the
UCRT's log1p is Goldberg's correction over its own log, u = 1 + x; log(u) - ((u - 1) - x) / u,
applied for every u (also above 2), and log1pf is the same statements in float32 over logf.
Log(double) / Log(float) ARE the UCRT's log / logf, so the model
equals NumPy on all 320,000 finite inputs per dtype of the probe (tiny and subnormal, small, (-1, 1), near -1, [1, 10],
up to the largest finite value), and the specials are pinned by hand: a NaN comes back untouched (a signalling NaN
stays signalling, sign and payload kept), log1p(-1) = -inf, log1p(+inf) = +inf, and every x < -1
(-inf included) gives the POSITIVE quiet NaN (.NET's Math.Log of a negative is the negative one).
Faithfully rounded, not correctly rounded: about 6% of float64 results sit one ULP from the true value, exactly where
NumPy's do.
expm1 / expm1f are bit-exact outside a band and within 2-3 ULP inside it. Outside
[-ln 2, ln 1.5], where |expm1(x)| >= 0.5 and the subtraction loses nothing, the UCRT returns exactly
exp(x) - 1 (expf(x) - 1f for float32, whose band ends at the float32 roundings of the same constants):
those results are NumPy's bit for bit. Inside the band the UCRT runs an undocumented approximation no public formula
reproduces (see Expm1Sun(double)), so the band runs Sun's s_expm1.c — accurate to one ULP of the true
value, equal to NumPy on about 58% of float64 and 57% of float32 in-band inputs, never more than 2 (float64) / 3
(float32) ULP from it. A NaN comes back untouched, -0.0 keeps its sign, expm1(-inf) = -1.
The naive forms these replace, Exp(x) - 1 and Log(1 + x) (also what .NET's ExpM1 /
LogP1 compute), cancel catastrophically near zero: expm1(1e-8) kept 8 digits, expm1(1e-300)
and log1p(1e-300) were 0, and -0.0 became +0.0. A platform pin, like NDHypotMath:
Linux NumPy calls glibc's routines, which round differently.
public static class NDLog1pExpm1Math
- Inheritance
-
NDLog1pExpm1Math
- Inherited Members
Methods
Expm1(double)
float64 np.expm1 of one element — e^x - 1 accurate near zero: NumPy's bits outside
[-ln 2, ln 1.5], Sun's s_expm1.c (within 2 ULP of NumPy) inside it.
public static double Expm1(double x)
Parameters
xdoubleThe argument.
Returns
- double
e^x - 1;-1for-inf(and oncee^xunderflows),+infon overflow, the sign of a zero kept, a NaN argument untouched.
Expm1F(float)
float32 np.expm1 of one element: the UCRT's expf(x) - 1f outside the float32 band (NumPy's bits),
the float64 Expm1Sun(double) rounded once inside it (within 3 ULP of NumPy's in-band polynomial, which is
itself up to 4 ULP from the true value).
public static float Expm1F(float x)
Parameters
xfloatThe argument.
Returns
- float
e^x - 1in float32, with Expm1(double)'s special values.
Log1p(double)
float64 np.log1p of one element — ln(1 + x) accurate near zero, NumPy's bits on win-amd64 (the UCRT
model of the class remarks).
public static double Log1p(double x)
Parameters
xdoubleThe argument.
Returns
- double
ln(1 + x);xitself while1 + xrounds to 1 (so-0.0, subnormals and tiny values return unchanged);-infat -1; the positive quiet NaN below -1; a NaN argument untouched.
Log1pF(float)
float32 np.log1p of one element — the UCRT's log1pf: Log1p(double)'s statements in
float32 over logf, NumPy's bits on win-amd64.
public static float Log1pF(float x)
Parameters
xfloatThe argument.
Returns
- float
ln(1 + x)in float32, with Log1p(double)'s special values (the positive quiet NaN is0x7FC00000).