Attention: Here be dragons
This is the latest
(unstable) version of this documentation, which may document features
not available in or compatible with released stable versions of NumDot.
Checking the stable version of the documentation...
ndi
Inherits: Object
Contains NumDot functions returning 64-bit signed integers.
Description
All mathematical functions in this namespace return 64-bit signed integers. If you're planning to use the result of reductions in further tensor computations, use nd instead. Using this namespace may unintentionally promote the prevalent data type of computations.
Methods
|
count_nonzero(a: |
|
max(a: |
|
median(a: |
|
min(a: |
|
norm(a: |
|
prod(a: |
|
sum(a: |
|
sum_product(a: |
|
trace(v: |
Method Descriptions
int count_nonzero(a: Variant) static 🔗
No-axis equivalent of nd.count_nonzero() that returns an int.
int max(a: Variant) static 🔗
No-axis equivalent of nd.max() that returns an int.
int median(a: Variant) static 🔗
No-axis equivalent of nd.median() that returns an int.
int min(a: Variant) static 🔗
No-axis equivalent of nd.min() that returns an int.
int norm(a: Variant, ord: Variant = 2) static 🔗
No-axis equivalent of nd.norm() that returns an int.
int prod(a: Variant) static 🔗
No-axis equivalent of nd.prod() that returns an int.
int sum(a: Variant) static 🔗
No-axis equivalent of nd.sum() that returns an int.
int sum_product(a: Variant, b: Variant) static 🔗
No-axis equivalent of nd.sum_product() that returns an int.
int trace(v: Variant, offset: int = 0, axis1: int = 0, axis2: int = 1) static 🔗
2-D only equivalent of nd.trace() that returns an int.