initial_pops

This function randomly initializes a population of the metaheuristic algorithm for a given number of repetitions.

initial_pops(n_repetitions, n_population, n_dimensions, x_lower, x_upper, type_pop, seeds)

Input variables

Name	Description	Type
`n_repetitions`	Number of repetitions	Integer
`n_population`	Number of population	Integer
`n_dimensions`	Problem dimension	Integer
`x_lower`	Lower limit of the design variables. Use None for combinatorial variables	List or None
`x_upper`	Upper limit of the design variables. Use None for combinatorial variables	List or None
`type_pop`	Type of population. Options: 'real code' or 'combinatorial code'. 'real code' call function initial_population_01 and 'combinatorial code' call function initial_population_02	String
`seeds`	Random seed. Use None for random seed	List or None
`population`	Population design variables. All repetitions	List

Output variables

Name	Description	Type
`population`	Population design variables. All repetitions	List

Example 1

Use the initial_pops function to randomly initialize the population required for four repetitions of the optimization procedure, where each solution set has two agents, and each has three dimensions. The design variables limits are \( \mathbf{x}_L = \left[1,\;1,\;1\right] \) and \( \mathbf{x}_U = \left[3,\;3,\;3\right] \). Use "seed without control" in your setup.

# Import
# pip install metapy-toolbox or pip install --upgrade metapy-toolbox
from metapy_toolbox import initial_pops # or import *

# Data
nRep = 4
nPop = 2
d = 3
xL = [1, 1, 1]
xU = [3, 3, 3]
typeCode = 'real code'
seeds = [None, None, None, None]

# Call function
pops = initial_pops(nRep, nPop, d, xL, xU, typeCode, seeds)

# Output details
print('population repetition ID = 0: ', pops[0])
print('population repetition ID = 1: ', pops[1])
print('population repetition ID = 2: ', pops[2])
print('population repetition ID = 3: ', pops[3])
print('\nAgent example:')
print('init. population rep. ID = 0 - pop = 0: ', pops[0][0])
print('init. population rep. ID = 0 - pop = 1: ', pops[0][1])

population repetition ID = 0:  [[1.7569870977744084, 1.042806656388939, 2.9574518690911633], [2.516871253402986, 1.0623716942485755, 2.1845178409922843]]
population repetition ID = 1:  [[2.1662762308582373, 2.66090910550537, 2.474471453333295], [2.8198256171594864, 2.570566076401062, 1.7930497331754311]]
population repetition ID = 2:  [[1.66330938269935, 1.870843964868756, 1.5196732590734792], [2.8295716904027115, 1.4195645627391247, 1.6252392995992408]]
population repetition ID = 3:  [[1.418822142851461, 2.083824683419757, 1.717224896458115], [2.795448903059947, 2.242610136700492, 1.1159414911258083]]

Agent example:
init. population rep. ID = 0 - pop = 0:  [1.7569870977744084, 1.042806656388939, 2.9574518690911633]
init. population rep. ID = 0 - pop = 1:  [2.516871253402986, 1.0623716942485755, 2.1845178409922843]

Example 2

# Import
# pip install metapy-toolbox or pip install --upgrade metapy-toolbox 
from metapy_toolbox import initial_pops # or import *

# Data
nRep = 4
nPop = 2
d = 3
xL = [1, 1, 1]
xU = [3, 3, 3]
typeCode = 'real code'
seeds = [10, 11, 12, 13]

# Call function
pops = initial_pops(nRep, nPop, d, xL, xU, typeCode, seeds)

# Output details
print('population repetition ID = 0: ', pops[0])
print('population repetition ID = 1: ', pops[1])
print('population repetition ID = 2: ', pops[2])
print('population repetition ID = 3: ', pops[3])
print('\nAgent example:')
print('init. population rep. ID = 0 - pop = 0: ', pops[0][0])
print('init. population rep. ID = 0 - pop = 1: ', pops[0][1])

population repetition ID = 0:  [[2.5426412865334918, 1.041503898718803, 2.2672964698525506], [2.4976077650772237, 1.9970140246051808, 1.4495932910616953]]
population repetition ID = 1:  [[1.3605393777535384, 1.0389504829752492, 1.9264370529966892], [2.4498678583842954, 1.8404072091754549, 1.9708541963355648]]
population repetition ID = 2:  [[1.3083256847593447, 2.4800993930308097, 1.5266300303702693], [2.0674787867605957, 1.0291499249708393, 2.83749401619977]]
population repetition ID = 3:  [[2.5554048211476403, 1.4750824400698246, 2.648557065322737], [2.9314983960859995, 2.9452022278097867, 1.9068984948346244]]

Agent example:
init. population rep. ID = 0 - pop = 0:  [2.5426412865334918, 1.041503898718803, 2.2672964698525506]
init. population rep. ID = 0 - pop = 1:  [2.4976077650772237, 1.9970140246051808, 1.4495932910616953]

Example 3

Use the initial_pops function to randomly initialize the population required for four repetitions of the optimization procedure, where each solution set has two agents, and each has three dimensions. Use the combinatorial process to generate an initial guess. Use "seed without control" in your setup.

# Import
# pip install metapy-toolbox or pip install --upgrade metapy-toolbox 
from metapy_toolbox import initial_pops # or import *

# Data
nRep = 4
nPop = 2
d = 10
xL = None
xU = None
typeCode = 'combinatorial code'
seeds = [None, None, None, None]

# Call function
pops = initial_pops(nRep, nPop, d, xL, xU, typeCode, seeds)

# Output details
print('population repetition ID = 0: ', pops[0])
print('population repetition ID = 1: ', pops[1])
print('population repetition ID = 2: ', pops[2])
print('population repetition ID = 3: ', pops[3])
print('\n Agent example:')
print('init. population rep. ID = 0 - pop = 0: ', pops[0][0])
print('init. population rep. ID = 0 - pop = 1: ', pops[0][1])

population repetition ID = 0:  [[3, 9, 0, 1, 8, 5, 7, 6, 2, 4], [4, 9, 2, 1, 7, 6, 8, 5, 3, 0]]
population repetition ID = 1:  [[7, 9, 0, 1, 3, 5, 4, 2, 6, 8], [9, 6, 4, 3, 2, 7, 0, 1, 5, 8]]
population repetition ID = 2:  [[1, 4, 2, 5, 0, 8, 6, 7, 9, 3], [8, 2, 7, 1, 0, 5, 3, 9, 4, 6]]
population repetition ID = 3:  [[6, 5, 0, 7, 8, 1, 2, 3, 4, 9], [4, 7, 9, 6, 3, 2, 0, 8, 1, 5]]

 Agent example:
init. population rep. ID = 0 - pop = 0:  [3, 9, 0, 1, 8, 5, 7, 6, 2, 4]
init. population rep. ID = 0 - pop = 1:  [4, 9, 2, 1, 7, 6, 8, 5, 3, 0]