Skip to content

objects.md

Latest commit

 

History

History
74 lines (60 loc) · 2.28 KB

objects.md

File metadata and controls

74 lines (60 loc) · 2.28 KB
Raw

OBJECTS

numpy

arange(int n) : array={0..n-1}
atleast_2d(array) :

  • for 1d array, adds empty dimension (useful for transpose)
    array copy(array) :
    clip(array, min, max) : return the array with values clipped in range
    diag(1-D_array, k=0) : matrix with 1-D_array on diagonal, k=which diagonal
  • diag(2-D_array) : vector with matrix’s diagonal
    dot() :
    array eye(int n, <int m>,) : identity matrix (ones on diagonal); n=cols, m=rows
    linspace(int start, int end, num=n) : returns n number at same distance between each other
    matmul(array1, array2) : product row by column
    max(array, axis=) :
    mean(array, axis=) : average
    min(array, axis=) :
    multiply(a1, a2) : mul one by one
    array ones() :
    outer() :
    prod() : product over axis
    range(max) : array of ints < max
    reshape(array) : keeps elements, changes shape
    std(matrix) : standard deviation
    sum(matrix, axis=) : sums, along axis
    transpose(matrix) :
  • for a 1d array, doesn't do anything;
  • do instead np.atleast_2d(A).T tril(matrix): triu(marix): upper diagonal unique(array, return_counts: bool): list of unique element, and counts if specified zeros(lunghezza(int)/dimensioni(tupla), tipo(default:float), qualcosa di memoria('C'/'F', default) ) -> array` : array full of zeroes

creation

meshgrid(x1:ndarray, x2, ...) : create grid, combining vectors; return x, y of coordinates

functions

cos(array:Array) : (like sin)
sin(array:Array) : compute sin for each point in array

TYPES

// float32
class array:

ATTRIBUTES

object : n-dimensional array (list/list of lists… or tuple for 1-dimension);
int size : number of elements (total)
tuple shape : dimensions
dtype : type of elements
int ndim : number of dimensions
T : transposed

METHODS

array(object, DTYPE=TYPE) : constructor

numpy.random

normal(loc=, scale=, size=) : random numbers with gauss distribution
rand(int n, int m) : random array, shape = n*m
randint(low, high=, size=, dtype=) : rand with ints, in range
seed(s:int) : set seed s

numpy.linalg (linear algebra)

cond() :
inv() : inverse
matrix_rank(matrix) :
norm() :

README.md