235 lines
6.7 KiB
Python
235 lines
6.7 KiB
Python
import sqlite3
|
|
from sqlite3 import Error
|
|
import os
|
|
|
|
|
|
class SQL_Database:
|
|
"""Main database class that controls all the functionality for input /
|
|
out of the database using SQLite3."""
|
|
|
|
sql_types_dict = [int, float, str]
|
|
|
|
def __init__(self):
|
|
self.conn = None
|
|
|
|
|
|
def create_connection(self, db_file):
|
|
"""Create a database connection to the SQLite database
|
|
specified by db_file."""
|
|
|
|
conn = None
|
|
try:
|
|
conn = sqlite3.connect(db_file)
|
|
return conn
|
|
except Error as e:
|
|
print(e)
|
|
|
|
self.conn = conn
|
|
|
|
|
|
def create_table(self, create_table_sql):
|
|
"""Create a table from the create_table_sql statement."""
|
|
|
|
try:
|
|
c = self.conn.cursor()
|
|
c.execute(create_table_sql)
|
|
except Error as e:
|
|
print(e)
|
|
|
|
|
|
def insert_chromosome(self, generation, chromosome):
|
|
""" Insert one chromosome into the database"""
|
|
|
|
# Structure the insert data
|
|
db_chromosome = (generation, chromosome.fitness, '[chromosome]')
|
|
|
|
# Create sql query structure
|
|
sql = ''' INSERT INTO data(generation,fitness,chromosome)
|
|
VALUES(?,?,?) '''
|
|
|
|
cur = self.conn.cursor()
|
|
cur.execute(sql, db_chromosome)
|
|
self.conn.commit()
|
|
return cur.lastrowid
|
|
|
|
|
|
def insert_current_population(self, ga):
|
|
""" Insert current generations population """
|
|
|
|
# Structure the insert data
|
|
db_chromosome_list = [
|
|
(ga.current_generation, chromosome.fitness, '[chromosome]')
|
|
for chromosome in ga.population.get_chromosome_list()
|
|
]
|
|
|
|
# Create sql query structure
|
|
sql = ''' INSERT INTO data(generation,fitness,chromosome)
|
|
VALUES(?,?,?) '''
|
|
|
|
cur = self.conn.cursor()
|
|
cur.executemany(sql, db_chromosome_list)
|
|
self.conn.commit()
|
|
return cur.lastrowid
|
|
|
|
|
|
def create_all_tables(self, ga):
|
|
"""Create the data table that store generation data."""
|
|
|
|
try:
|
|
# Remove old database file if it exists.
|
|
os.remove(ga.database_name)
|
|
except:
|
|
# If the database does not exist continue
|
|
pass
|
|
|
|
# create a database connection
|
|
self.conn = self.create_connection(ga.database_name)
|
|
|
|
# create tables
|
|
if self.conn is not None:
|
|
# create projects table
|
|
self.create_table(ga.sql_create_data_structure)
|
|
self.create_table(ga.sql_create_config_structure)
|
|
else:
|
|
print("Error! cannot create the database connection.")
|
|
|
|
|
|
def insert_config(self,ga):
|
|
"""Insert the configuration attributes into the config."""
|
|
|
|
# Structure the insert data
|
|
db_config_list = [
|
|
ga.chromosome_length,
|
|
ga.population_size,
|
|
ga.chromosome_impl,
|
|
ga.gene_impl,
|
|
ga.target_fitness_type,
|
|
ga.update_fitness,
|
|
ga.parent_ratio,
|
|
ga.selection_probability,
|
|
ga.tournament_size_ratio,
|
|
ga.current_generation,
|
|
float(ga.current_fitness),
|
|
ga.generation_goal,
|
|
ga.fitness_goal,
|
|
ga.tolerance_goal,
|
|
ga.percent_converged,
|
|
ga.chromosome_mutation_rate,
|
|
ga.gene_mutation_rate,
|
|
ga.initialization_impl,
|
|
ga.fitness_function_impl,
|
|
ga.parent_selection_impl,
|
|
ga.crossover_individual_impl,
|
|
ga.crossover_population_impl,
|
|
ga.survivor_selection_impl,
|
|
ga.mutation_individual_impl,
|
|
ga.mutation_population_impl,
|
|
ga.termination_impl,
|
|
ga.database_name
|
|
]
|
|
|
|
# Clean up so the sqlite database accepts the data structure
|
|
for i in range(len(db_config_list)):
|
|
if callable(db_config_list[i]):
|
|
db_config_list[i] = db_config_list[i].__name__
|
|
elif type(db_config_list[i]) not in self.sql_types_dict:
|
|
db_config_list[i] = str(db_config_list[i])
|
|
|
|
# For some reason it has to be in var = array(tuple()) form
|
|
db_config_list = [tuple(db_config_list)]
|
|
|
|
# Create sql query structure
|
|
sql = f'''INSERT INTO config (chromosome_length,
|
|
population_size,
|
|
chromosome_impl,
|
|
gene_impl,
|
|
target_fitness_type,
|
|
update_fitness,
|
|
parent_ratio,
|
|
selection_probability,
|
|
tournament_size_ratio,
|
|
current_generation,
|
|
current_fitness,
|
|
generation_goal,
|
|
fitness_goal,
|
|
tolerance_goal,
|
|
percent_converged,
|
|
chromosome_mutation_rate,
|
|
gene_mutation_rate,
|
|
initialization_impl,
|
|
fitness_function_impl,
|
|
parent_selection_impl,
|
|
crossover_individual_impl,
|
|
crossover_population_impl,
|
|
survivor_selection_impl,
|
|
mutation_individual_impl,
|
|
mutation_population_impl,
|
|
termination_impl,
|
|
database_name) VALUES({(',?'*len(db_config_list))[1:]}) '''
|
|
|
|
# Execute sql query
|
|
cur = self.conn.cursor()
|
|
cur.executemany(sql, db_config_list)
|
|
self.conn.commit()
|
|
return cur.lastrowid
|
|
|
|
|
|
def query_all(self, query):
|
|
"""Query for muliple rows of data"""
|
|
|
|
cur = self.conn.cursor()
|
|
cur.execute(query)
|
|
|
|
return cur.fetchall()
|
|
|
|
|
|
def query_one_item(self, query):
|
|
"""Query for single data point"""
|
|
|
|
cur = self.conn.cursor()
|
|
cur.execute(query)
|
|
query_data = cur.fetchone()
|
|
|
|
return query_data[0]
|
|
|
|
|
|
def get_generation_total_fitness(self):
|
|
"""Get each generations total fitness sum from the database """
|
|
|
|
query_data = self.query_all("SELECT SUM(fitness) FROM data GROUP BY generation;")
|
|
|
|
# Format the fitness data into one list
|
|
formated_query_data = [i[0] for i in query_data]
|
|
|
|
return formated_query_data
|
|
|
|
|
|
def get_total_generations(self):
|
|
"""Get the total generations from the database"""
|
|
|
|
query_data = self.query_one_item("SELECT COUNT(DISTINCT generation) FROM data;")
|
|
|
|
return query_data
|
|
|
|
|
|
def get_highest_chromosome(self):
|
|
"""Get the highest fitness of each generation"""
|
|
|
|
query_data = self.query_all("select fitness, max(fitness) from data group by generation")
|
|
|
|
# Format the fitness data into one list
|
|
formated_query_data = [i[0] for i in query_data]
|
|
|
|
return formated_query_data;
|
|
|
|
|
|
def get_lowest_chromosome(self):
|
|
"""Get the lowest fitness of each generation"""
|
|
|
|
query_data = self.query_all("select fitness, min(fitness) from data group by generation")
|
|
|
|
# Format the fitness data into one list
|
|
formated_query_data = [i[0] for i in query_data]
|
|
|
|
return formated_query_data;
|