true-false-python

PyPI package with some better syntax tools for python

pip install truefalsepython

• true-false-python
  • Logical tools
  • Little functions
  • Useful functions
  • R-like functions
  • Functions without category

Logical tools

For True and False values there are equal constants (like it is in C-like languages or R) TRUE, T, true and FALSE, F, false; for None there is NULL constant:

from truefalsepython import TRUE, FALSE, T, F, true, false, NULL

print(True == T) # True
print(True == TRUE) # True
print(True == true) # True


print(False == F) # True
print(False == FALSE) # True
print(False == false) # True

print(NULL) # None

Little functions

• is_odd(number)
• is_even(number)
• is_number(object)
• max_fast(a, b)
• min_fast(a, b)

Useful functions

• fast_sample(objects, probs) -- returns 1 random object from objects with probs probabilities. It's faster than np.random.choice(objects, 1, probs) (example)
• randomTrue(prob = 0.5) -- returns True with probability prob, otherwise False

Useful for debug:

• set_trace() -- like breakpoint
• debug(function, *args, **kwargs) -- for debug function function with those arguments

R-like functions

For arrays there are several R-like functions:

• ifelse — just wrapper of numpy.where
• nrow — returns number of rows
• ncol — returns number of columns
• colMeans — returns average for each column
• rowMeans — returns average for each row
• colSums — returns sums for each column
• rowSums — returns sums for each row
• apply — applies function FUN to dimension of arr2D array (for rows if MARGIN == 1 and columns if MARGIN == 2)
• lapply — applies function func for each element in array (array/list or something else)
• sapply — like lapply but returns numpy array
• sample — it is np.random.choice but replace = False by default
• sample_int — sample numbers from 0 to n-1

Example of usage:

import numpy as np
from truefalsepython import nrow, ncol, colMeans, rowMeans, colSums, rowSums, apply, lapply, sapply, sample, sample_int


np.random.seed(1)

# some 2D array
random_matrix = np.random.randint(8, size = (5, 3))


# how to get rows and cols counts
print(nrow(random_matrix)) # 5

print(ncol(random_matrix)) # 3

# operations for each row/column
print(rowMeans(random_matrix))
# [4.         2.66666667 5.         0.33333333 5.33333333]

print(colMeans(random_matrix))
# [2.4 4.4 3.6]

print(rowSums(random_matrix))
# [12  8 15  1 16]

print(colSums(random_matrix))
# [12 22 18]

# apply function (MARGIN is 1 for rows and 2 for columns)
print(apply(random_matrix, MARGIN = 1, FUN = np.min))
# [3 0 3 0 4]

# as u can see, it's not necessary to use FUN returns only 1 number by vector
print(apply(random_matrix, MARGIN = 2, FUN = np.sqrt))
#[[2.23606798 0.         1.73205081 0.         2.        ]
# [1.73205081 2.64575131 2.23606798 0.         2.64575131]
# [2.         1.         2.64575131 1.         2.23606798]]

some_arr = np.array([1, 2, 3, 5, 4, 3, 2])

# returns list
print(lapply(some_arr, lambda x: -x))
# [-1, -2, -3, -5, -4, -3, -2]

# returns numpy array
print(sapply(some_arr, lambda x: -x))
# [-1 -2 -3 -5 -4 -3 -2]

# like np.random.choice but replace = False by default
print(sample(some_arr, 4))
# [5 3 2 1]

# sample numbers from 0 to n-1
print(sample_int(n = 100, size = 10))
# [69 46 58 12 73 98 31 53 65 96]

Functions without category

• time_to_seconds(days = 0, hours = 0, minutes = 0, seconds = 5) -- converts time to seconds