Skip to content

Latest commit

 

History

History
159 lines (140 loc) · 3.71 KB

README.md

File metadata and controls

159 lines (140 loc) · 3.71 KB

numpy module for skulpt

##Description This is a partial port of some numpy functions for skulpt. Mainly, focused on ndarray and its respective functions. See below for the currently supported functions.

News

Latest

  • improved internal attribute handling
  • improved internal buffer handling
  • added ndarray.T attribute
  • added __getattr__
  • added __setattr__
  • added numpy.vdot

Example:

import numpy as np

a = np.array([[1, 2], [3, 4]])
print(a)
a.shape = (4,)
print(a)
a.shape = (1, 2, 2)
print(a.T)
a.shape = 4
print(a)

a = np.array([[1j, 2j], [3j, 4j]])
b = [1, 2, 3, 4]
c = np.vdot(a, b)

print(c)

##Supported

###ndarray Currently all attributes are readonly. Though

  • __str__
  • __repr__
  • __len__
  • shape
  • ndim
  • data
  • dtype
  • size
  • tolist
  • fill
  • [], with slices
  • operators: +, -, /, *, **, %, neg, pos, xor, shift
  • reshape
  • copy
  • __iter__

###functions

  • linspace
  • array
  • ones
  • zeros
  • full
  • empty
  • arange
  • asarray

###trigonomeric Some of the functions rely on math.js, but you can use the module without it. All functions that require the library are going to throw an python exception if they cannot call mathjs.

  • dot (mathjs)
  • sin
  • cos
  • tan
  • sinh (mathjs)
  • cosh (mathjs)
  • tanh (mathjs)
  • arctan
  • arcsin
  • arccos

###Examples

import numpy as np

a = [[1, 2], [3, 4]]

b = np.array(a, ndmin=3, dtype=float)
print "np.array(a, ndmin=3a, dtype=float)"

c = np.ones(b.shape)
print "np.ones(b.shape): %s" % (c,)
d = np.zeros(b.shape)
print "np.zeros(b.shape): %s" % (d,)
print "__str__: %s" % b
print "__repr__: %r" % b
print "__len__: %s" % len(b)
print "shape %s" % (b.shape,)
print "ndim %s" % b.ndmin
print "data: %s"  % (b.data,)
print "dtype: %s" % b.dtype
print "size %s" % b.size
print "b.tolist %s" % (b.tolist(),)
b.fill(9)
print "b.fill(9): %s" % (b,)
b[0, 0, 0] = 2
print "b[0, 0, 0] = 2: %s" % (b,)

print ""
print "np.full((2,2), 2.0)"
c = np.full((2,2), 2.0, int)

print "===================="
print "     operations"
print "===================="
print "c = %s" % (c,)
print "c + 2 = %s" % (c + 2,)
print "c - 2 = %s" % (c - 2,)
print "c * 2 = %s" % (c * 2,)
print "c / 2 = %s" % (c / 2,)
print "c ** 2 = %s" % (c ** 2,)
print "+c = %s" % (+c,)
print "-c = %s" % (-c,)

print "===================="
print "   trigonometric    "
print "===================="
c = np.full((2,2), 0, int)
print "c = %s" % (c,)
print "np.sin(c) = %s" % (np.sin(c),)
print "np.cos(c) = %s" % (np.cos(c),)
print "np.tan(c) = %s" % (np.tan(c),)
print "np.arcsin(c) = %s" % (np.arcsin(c),)
print "np.arccos(c) = %s" % (np.arccos(c),)
print "np.arctan(c) = %s" % (np.arctan(c),)
print "np.sinh(c) = %s" % (np.sinh(c),)
print "np.cosh(c) = %s" % (np.cosh(c),)
print "np.tanh(c) = %s" % (np.tanh(c),)
print "np.sin([0,1]) = %s" % (np.sin([0,1]),)
print "np.sin((0,1)) = %s" % (np.sin((0,1)),)

print "===================="
print "      various       "
print "===================="
ar = np.arange(3.0)
print "np.arange(3.0): %s, dtype: %s" % (ar, ar.dtype)
ar = np.arange(3)
print "np.arange(3): %s, dtype: %s" % (ar, ar.dtype)
ar = np.arange(3,7)
print "np.arange(3,7): %s, dtype: %s" % (ar, ar.dtype)
ar = np.arange(3,7, 2)
print "np.arange(3,7, 2): %s, dtype: %s" % (ar, ar.dtype)
ar = np.linspace(2.0, 3.0, num=5)
print "np.linspace(2.0, 3.0, num=5): %s" % (ar,)
ar = np.linspace(2.0, 3.0, num=5, endpoint=False)
print "np.linspace(2.0, 3.0, num=5, endpoint=False): %s" % (ar,)
ar = np.linspace(2.0, 3.0, num=5, retstep=True)
print "np.linspace(2.0, 3.0, num=5, retstep=True): %s" % (ar,)