Permalink
Browse files

first commit

  • Loading branch information...
tinybike committed Dec 6, 2014
0 parents commit a5c001b2266339b52d1a1807901a57d1a7ef7ba4
Showing with 952 additions and 0 deletions.
  1. +53 −0 .gitignore
  2. +2 −0 README.md
  3. +178 −0 cumulants.py
  4. +119 −0 fixedpoint.se
  5. +120 −0 test_cumulants.py
  6. +17 −0 test_math.py
  7. +463 −0 truthcoin.se
@@ -0,0 +1,53 @@
# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]

# C extensions
*.so

# Distribution / packaging
.Python
env/
build/
develop-eggs/
dist/
eggs/
lib/
lib64/
parts/
sdist/
var/
*.egg-info/
.installed.cfg
*.egg

# PyInstaller
# Usually these files are written by a python script from a template
# before PyInstaller builds the exe, so as to inject date/other infos into it.
*.manifest
*.spec

# Installer logs
pip-log.txt
pip-delete-this-directory.txt

# Unit test / coverage reports
htmlcov/
.tox/
.coverage
.cache
nosetests.xml
coverage.xml

# Translations
*.mo
*.pot

# Django stuff:
*.log

# Sphinx documentation
docs/_build/

# PyBuilder
target/
@@ -0,0 +1,2 @@
serpent-core
------------
@@ -0,0 +1,178 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
"""
Cumulant tensors and statistics. Semi-compatible with Serpent.
"""
from __future__ import division
import array as arr

def array(size, typecode='f'):
# Emulates Serpent arrays.
#
# Args:
# size (int): number of elements in the array
# typecode (char): data type the array will hold (default: float)
#
a = arr.array(typecode, (0,)*size)
return(a)

def mean(u, size):
# Calculates the arithmetic mean.
#
# Args:
# u: numeric array (vector)
# size (int): number of elements in u
#
m = 0
while i < size:
m += u[i]
m /= size
return(m)

def dot(u, v, size):
# Calculates the dot (inner) product.
#
# Args:
# u: numeric array (vector)
# v: numeric array (vector)
# size (int): number of elements in u
#
prod = 0
i = 0
while i < size:
prod += u[i] * v[i]
i += 1
return(prod)

def cov(data, rows, cols, unbias):
# Covariance matrix (second cumulant).
#
# Args:
# data: two-dimensional data matrix (signals = columns, samples = rows)
# rows: number of rows (samples per signal) in the data matrix
# cols: number of columns (signals) in the data matrix
#
tensor = [[]] * cols
i = 0
while i < cols:
j = 0
tensor[i] = array(cols)
while j < cols:
u = 0
row = 0
while row < rows:
i_mean = 0
j_mean = 0
r = 0
while r < rows:
i_mean += data[r][i]
j_mean += data[r][j]
r += 1
i_mean /= rows
j_mean /= rows
i_center = data[row][i] - i_mean
j_center = data[row][j] - j_mean
u += i_center * j_center
row += 1
tensor[i][j] = u / (rows - unbias)
j += 1
i += 1
return tensor

def coskew(data, rows, cols, unbias):
# Block-unfolded third cumulant tensor.
#
# Args:
# data: two-dimensional data matrix (signals = columns, samples = rows)
# rows: number of rows (samples per signal) in the data matrix
# cols: number of columns (signals) in the data matrix
#
tensor = [[]] * cols
k = 0
while k < cols:
face = [[]] * cols
i = 0
while i < cols:
j = 0
face[i] = array(cols)
while j < cols:
u = 0
row = 0
while row < rows:
i_mean = 0
j_mean = 0
k_mean = 0
r = 0
while r < rows:
i_mean += data[r][i]
j_mean += data[r][j]
k_mean += data[r][k]
r += 1
i_mean /= rows
j_mean /= rows
k_mean /= rows
i_center = data[row][i] - i_mean
j_center = data[row][j] - j_mean
k_center = data[row][k] - k_mean
u += i_center * j_center * k_center
row += 1
face[i][j] = u / (rows - unbias)
j += 1
tensor[k] = face
i += 1
k += 1
return tensor

def cokurt(data, rows, cols, unbias):
# Block-unfolded fourth cumulant tensor.
#
# Args:
# data: two-dimensional data matrix (signals = columns, samples = rows)
# rows: number of rows (samples per signal) in the data matrix
# cols: number of columns (signals) in the data matrix
#
tensor = [[]] * cols
l = 0
while l < cols:
block = [[]] * cols
k = 0
while k < cols:
face = [[]] * cols
i = 0
while i < cols:
j = 0
face[i] = array(cols)
while j < cols:
u = 0
row = 0
while row < rows:
i_mean = 0
j_mean = 0
k_mean = 0
l_mean = 0
r = 0
while r < rows:
i_mean += data[r][i]
j_mean += data[r][j]
k_mean += data[r][k]
l_mean += data[r][l]
r += 1
i_mean /= rows
j_mean /= rows
k_mean /= rows
l_mean /= rows
i_center = data[row][i] - i_mean
j_center = data[row][j] - j_mean
k_center = data[row][k] - k_mean
l_center = data[row][l] - l_mean
u += i_center * j_center * k_center * l_center
row += 1
face[i][j] = u / (rows - unbias)
j += 1
block[k] = face
i += 1
tensor[l] = block
k += 1
l += 1
return tensor
Oops, something went wrong.

0 comments on commit a5c001b

Please sign in to comment.