Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

rename to PEP 8 #151

Merged
merged 5 commits into from
Jul 10, 2023
Merged

rename to PEP 8 #151

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
0cc3d7b
rename 1
kozinove Jul 7, 2023
3804a55
rename 2
kozinove Jul 7, 2023
387f23e
rename 4
kozinove Jul 7, 2023
4c21be8
rename 6
kozinove Jul 7, 2023
b53871f
rename 6
kozinove Jul 7, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
39 changes: 19 additions & 20 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -92,27 +92,26 @@ if __name__ == "__main__":
"""
Minimization of the Rastrigin test function with visualization
"""
#Create a test task
# Create a test task
problem = Rastrigin(2)
#Setup a solver options
params = SolverParameters(r=2.5, eps=0.01, itersLimit=300, refineSolution=True)
#Create the solver
# Setup a solver options
params = SolverParameters(r=2.5, eps=0.01, iters_limit=300, refine_solution=True)
# Create the solver
solver = Solver(problem, parameters=params)
#Print results to console while solving
# Print results to console while solving
cfol = ConsoleOutputListener(mode='full')
solver.AddListener(cfol)
#3D visualization at the end of the solution
spl = StaticPainterNDListener("rastrigin.png", "output", varsIndxs=[0,1], mode="surface", calc="interpolation")
solver.AddListener(spl)
#Run problem solution
sol = solver.Solve()
solver.add_listener(cfol)
# 3D visualization at the end of the solution
spl = StaticPainterNDListener("rastrigin.png", "output", vars_indxs=[0, 1], mode="surface", calc="interpolation")
solver.add_listener(spl)
# Run problem solution
sol = solver.solve()
```

# **Examples**

Let’s demonstrate the use of the iOpt framework when tuning the hyperparameters of one of the machine learning methods. In the support vector machine ([SVC](https://scikit-learn.org/stable/modules/generated/sklearn.svm.SVC.html)), we find the optimal hyperparameters (the regularization parameter **C**, the kernel coefficient **gamma**) in the problem of breast cancer classification ([detailed description of the data](https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic))).


```python
import numpy as np
from sklearn.utils import shuffle
Expand Down Expand Up @@ -140,19 +139,19 @@ if __name__ == "__main__":

problem = SVC_2d.SVC_2D(x, y, regularization_value_bound, kernel_coefficient_bound)

method_params = SolverParameters(r=np.double(3.0), itersLimit=100)
method_params = SolverParameters(r=np.double(3.0), iters_limit=100)
solver = Solver(problem, parameters=method_params)

apl = AnimatePainterNDListener("svc2d_anim.png", "output", varsIndxs=[0, 1], toPaintObjFunc=False)
solver.AddListener(apl)
apl = AnimatePainterNDListener("svc2d_anim.png", "output", vars_indxs=[0, 1], to_paint_obj_func=False)
solver.add_listener(apl)

spl = StaticPainterNDListener("svc2d_stat.png", "output", vars_indxs=[0, 1], mode="surface", calc="interpolation")
solver.add_listener(spl)

spl = StaticPainterNDListener("svc2d_stat.png", "output", varsIndxs=[0, 1], mode="surface", calc="interpolation")
solver.AddListener(spl)

cfol = ConsoleOutputListener(mode='full')
solver.AddListener(cfol)
solver.add_listener(cfol)

solver_info = solver.Solve()
solver_info = solver.solve()

```

Expand Down
71 changes: 35 additions & 36 deletions README_ru.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,23 +64,23 @@ from iOpt.method.listener import StaticNDPaintListener, ConsoleFullOutputListene
from subprocess import Popen, PIPE, STDOUT

if __name__ == "__main__":
"""
Минимизация функции Растригина и визуализация
"""
#Создание тестовой задачи
problem = Rastrigin(2)
#Установка параметров поиска оптимального решения
params = SolverParameters(r=2.5, eps=0.01, itersLimit=300, refineSolution=True)
#Создание решателя
solver = Solver(problem, parameters=params)
#Вывод результатов на консоль в процессе решения
cfol = ConsoleFullOutputListener(mode='full')
solver.AddListener(cfol)
#3D визуализация по окончании решения
spl = StaticNDPaintListener("rastrigin.png", "output", varsIndxs=[0,1], mode="surface", calc="interpolation")
solver.AddListener(spl)
#Запуск решения задачи
sol = solver.Solve()
"""
Минимизация функции Растригина и визуализация
"""
# Создание тестовой задачи
problem = Rastrigin(2)
# Установка параметров поиска оптимального решения
params = SolverParameters(r=2.5, eps=0.01, iters_limit=300, refine_solution=True)
# Создание решателя
solver = Solver(problem, parameters=params)
# Вывод результатов на консоль в процессе решения
cfol = ConsoleFullOutputListener(mode='full')
solver.add_listener(cfol)
# 3D визуализация по окончании решения
spl = StaticNDPaintListener("rastrigin.png", "output", vars_indxs=[0, 1], mode="surface", calc="interpolation")
solver.add_listener(spl)
# Запуск решения задачи
sol = solver.solve()
```

# **Примеры**
Expand All @@ -90,7 +90,6 @@ if __name__ == "__main__":
вещественные гиперпараметры (**C** - параметр регуляризации, **gamma** - коэффициент ядра) для решения задачи классификации рака молочной железы
([подробное описание данных](https://archive.ics.uci.edu/ml/datasets/Breast+Cancer+Wisconsin+(Diagnostic))).


```python
import numpy as np
from sklearn.utils import shuffle
Expand All @@ -103,32 +102,32 @@ from examples.Machine_learning.SVC._2D.Problems import SVC_2d


def load_breast_cancer_data():
dataset = load_breast_cancer()
x_raw, y_raw = dataset['data'], dataset['target']
inputs, outputs = shuffle(x_raw, y_raw ^ 1, random_state=42)
return inputs, outputs
dataset = load_breast_cancer()
x_raw, y_raw = dataset['data'], dataset['target']
inputs, outputs = shuffle(x_raw, y_raw ^ 1, random_state=42)
return inputs, outputs


if __name__ == "__main__":
x, y = load_breast_cancer_data()
regularization_value_bound = {'low': 1, 'up': 6}
kernel_coefficient_bound = {'low': -7, 'up': -3}
x, y = load_breast_cancer_data()
regularization_value_bound = {'low': 1, 'up': 6}
kernel_coefficient_bound = {'low': -7, 'up': -3}

problem = SVC_2d.SVC_2D(x, y, regularization_value_bound, kernel_coefficient_bound)

problem = SVC_2d.SVC_2D(x, y, regularization_value_bound, kernel_coefficient_bound)
method_params = SolverParameters(r=np.double(3.0), iters_limit=100)
solver = Solver(problem, parameters=method_params)

method_params = SolverParameters(r=np.double(3.0), itersLimit=100)
solver = Solver(problem, parameters=method_params)
apl = AnimationNDPaintListener("svc2d_anim.png", "output", vars_indxs=[0, 1], to_paint_obj_func=False)
solver.add_listener(apl)

apl = AnimationNDPaintListener("svc2d_anim.png", "output", varsIndxs=[0, 1], toPaintObjFunc=False)
solver.AddListener(apl)
spl = StaticNDPaintListener("svc2d_stat.png", "output", vars_indxs=[0, 1], mode="surface", calc="interpolation")
solver.add_listener(spl)

spl = StaticNDPaintListener("svc2d_stat.png", "output", varsIndxs=[0, 1], mode="surface", calc="interpolation")
solver.AddListener(spl)

cfol = ConsoleFullOutputListener(mode='full')
solver.AddListener(cfol)
cfol = ConsoleFullOutputListener(mode='full')
solver.add_listener(cfol)

solver_info = solver.Solve()
solver_info = solver.solve()

```

Expand Down
92 changes: 46 additions & 46 deletions docs/source/examples.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -93,8 +93,8 @@ ____________________________________________________________________________
import xml.etree.ElementTree as ET
import numpy as np

def load_TSPs_matrix(filename):
root = ET.parse(filename).getroot()
def load_TSPs_matrix(file_name):
root = ET.parse(file_name).getroot()
columns = root.findall('graph/vertex')
num_cols = len(columns)
trans_matrix = np.zeros((num_cols, num_cols))
Expand Down Expand Up @@ -139,8 +139,8 @@ ____________________________________________________________________________
import numpy as np
from sko.GA import GA_TSP

def load_TSPs_matrix(filename):
root = ET.parse(filename).getroot()
def load_TSPs_matrix(file_name):
root = ET.parse(file_name).getroot()
columns = root.findall('graph/vertex')
num_cols = len(columns)
trans_matrix = np.zeros((num_cols, num_cols))
Expand Down Expand Up @@ -189,22 +189,22 @@ ____________________________________________________________________________
population_size: int,
mutation_probability_bound: Dict[str, float]):
self.dimension = 1
self.numberOfFloatVariables = 1
self.numberOfDiscreteVariables = 0
self.numberOfObjectives = 1
self.numberOfConstraints = 0
self.number_of_float_variables = 1
self.number_of_discrete_variables = 0
self.number_of_objectives = 1
self.number_of_constraints = 0
self.costMatrix = cost_matrix
if num_iteration <= 0:
raise ValueError('The number of iterations cannot be zero or negative.')
if population_size <= 0:
raise ValueError('Population size cannot be negative or zero')
self.populationSize = population_size
self.numberOfIterations = num_iteration
self.floatVariableNames = np.array(["Mutation probability"],
self.float_variable_names = np.array(["Mutation probability"],
dtype=str)
self.lowerBoundOfFloatVariables =
self.lower_bound_of_float_variables =
np.array([mutation_probability_bound['low']], dtype=np.double)
self.upperBoundOfFloatVariables =
self.upper_bound_of_float_variables =
np.array([mutation_probability_bound['up']], dtype=np.double)
self.n_dim = cost_matrix.shape[0]

Expand All @@ -215,7 +215,7 @@ ____________________________________________________________________________

def Calculate(self, point: Point,
functionValue: FunctionValue) -> FunctionValue:
mutation_prob = point.floatVariables[0]
mutation_prob = point.float_variables[0]
ga_tsp = GA_TSP(func=self.calc_total_distance,
n_dim=self.n_dim, size_pop=self.populationSize,
max_iter=self.numberOfIterations,
Expand All @@ -240,8 +240,8 @@ ____________________________________________________________________________
import numpy as np
import xml.etree.ElementTree as ET

def load_TSPs_matrix(filename):
root = ET.parse(filename).getroot()
def load_TSPs_matrix(file_name):
root = ET.parse(file_name).getroot()
columns = root.findall('graph/vertex')
num_cols = len(columns)
trans_matrix = np.zeros((num_cols, num_cols))
Expand All @@ -258,7 +258,7 @@ ____________________________________________________________________________
mutation_probability_bound = {'low': 0.0, 'up': 1.0}
problem = ga_tsp_vary_mutation.GA_TSP_Vary_Mutation(tsp_matrix,
num_iteration, population_size, mutation_probability_bound)
method_params = SolverParameters(r=np.double(3.0), itersLimit=20)
method_params = SolverParameters(r=np.double(3.0), iters_limit=20)
solver = Solver(problem, parameters=method_params)

solver_info = solver.Solve()
Expand Down Expand Up @@ -397,26 +397,26 @@ ________________________________________________________________________________
kernel_coefficient_bound: Dict[str, float]):

self.dimension = 2
self.numberOfFloatVariables = 2
self.numberOfDiscreteVariables = 0
self.numberOfObjectives = 1
self.numberOfConstraints = 0
self.number_of_float_variables = 2
self.number_of_discrete_variables = 0
self.number_of_objectives = 1
self.number_of_constraints = 0
if x_dataset.shape[0] != y_dataset.shape[0]:
raise ValueError('The input and output sample sizes do not match.')
self.x = x_dataset
self.y = y_dataset
self.floatVariableNames = np.array(["Regularization parameter",
self.float_variable_names = np.array(["Regularization parameter",
"Kernel coefficient"], dtype=str)
self.lowerBoundOfFloatVariables =
self.lower_bound_of_float_variables =
np.array([regularization_bound['low'],
kernel_coefficient_bound['low']], dtype=np.double)
self.upperBoundOfFloatVariables =
self.upper_bound_of_float_variables =
np.array([regularization_bound['up'],
kernel_coefficient_bound['up']], dtype=np.double)

def Calculate(self, point: Point,
functionValue: FunctionValue) -> FunctionValue:
cs, gammas = point.floatVariables[0], point.floatVariables[1]
cs, gammas = point.float_variables[0], point.float_variables[1]
clf = SVC(C=10**cs, gamma=10**gammas)
clf.fit(self.x, self.y)
functionValue.value = -cross_val_score(clf, self.x, self.y,
Expand Down Expand Up @@ -456,24 +456,24 @@ ________________________________________________________________________________
problem = SVC_2d.SVC_2D(x, y, regularization_value_bound,
kernel_coefficient_bound)

method_params = SolverParameters(r=np.double(3.0), itersLimit=10)
method_params = SolverParameters(r=np.double(3.0), iters_limit=10)
solver = Solver(problem, parameters=method_params)

apl = AnimationNDPaintListener("svc2d_anim.png", "output",
varsIndxs=[0, 1], toPaintObjFunc=False)
vars_indxs=[0, 1], to_paint_obj_func=False)
solver.AddListener(apl)

spl = StaticNDPaintListener("svc2d_stat.png", "output", varsIndxs=[0, 1],
spl = StaticNDPaintListener("svc2d_stat.png", "output", vars_indxs=[0, 1],
mode="surface", calc="interpolation")
solver.AddListener(spl)

solver_info = solver.Solve()
print(solver_info.numberOfGlobalTrials)
print(solver_info.numberOfLocalTrials)
print(solver_info.solvingTime)
print(solver_info.number_of_global_trials)
print(solver_info.number_of_local_trials)
print(solver_info.solving_time)

print(solver_info.bestTrials[0].point.floatVariables)
print(solver_info.bestTrials[0].functionValues[0].value)
print(solver_info.best_trials[0].point.float_variables)
print(solver_info.best_trials[0].function_values[0].value)

После проведения эксперимента программа выводит общее время поиска оптимума, точку на сетке, в которой достигается
оптимум, найденное максимальное значение метрики f1-score, а также интерполирует график целевой функции.
Expand Down Expand Up @@ -546,25 +546,25 @@ ________________________________________________________________________________
):
super(SVC_3D, self).__init__()
self.dimension = 3
self.numberOfFloatVariables = 2
self.numberOfDiscreteVariables = 1
self.numberOfObjectives = 1
self.numberOfConstraints = 0
self.number_of_float_variables = 2
self.number_of_discrete_variables = 1
self.number_of_objectives = 1
self.number_of_constraints = 0
if x_dataset.shape[0] != y_dataset.shape[0]:
raise ValueError('The input and output sample sizes do not match.')
self.x = x_dataset
self.y = y_dataset
self.floatVariableNames = np.array(["Regularization parameter", "Kernel coefficient"], dtype=str)
self.lowerBoundOfFloatVariables = np.array([regularization_bound['low'], kernel_coefficient_bound['low']],
self.float_variable_names = np.array(["Regularization parameter", "Kernel coefficient"], dtype=str)
self.lower_bound_of_float_variables = np.array([regularization_bound['low'], kernel_coefficient_bound['low']],
dtype=np.double)
self.upperBoundOfFloatVariables = np.array([regularization_bound['up'], kernel_coefficient_bound['up']],
self.upper_bound_of_float_variables = np.array([regularization_bound['up'], kernel_coefficient_bound['up']],
dtype=np.double)
self.discreteVariableNames.append('kernel')
self.discreteVariableValues.append(kernel_type['kernel'])
self.discrete_variable_names.append('kernel')
self.discrete_variable_values.append(kernel_type['kernel'])

def Calculate(self, point: Point, functionValue: FunctionValue) -> FunctionValue:
cs, gammas = point.floatVariables[0], point.floatVariables[1]
kernel_type = point.discreteVariables[0]
cs, gammas = point.float_variables[0], point.float_variables[1]
kernel_type = point.discrete_variables[0]
clf = SVC(C=10 ** cs, gamma=10 ** gammas, kernel=kernel_type)
functionValue.value = -cross_val_score(clf, self.x, self.y, scoring='f1').mean()
return functionValue
Expand Down Expand Up @@ -641,7 +641,7 @@ ________________________________________________________________________________
kernel_coefficient_bound = {'low': -9, 'up': -6.7}
kernel_type = {'kernel': ['rbf', 'sigmoid', 'poly']}
problem = SVC_3D.SVC_3D(x, y, regularization_value_bound, kernel_coefficient_bound, kernel_type)
method_params = SolverParameters(itersLimit=400)
method_params = SolverParameters(iters_limit=400)
solver = Solver(problem, parameters=method_params)
apl = StaticDiscreteListener("experiment1.png", mode='analysis')
solver.AddListener(apl)
Expand Down Expand Up @@ -794,11 +794,11 @@ ________________________________________________________________________________
regularization_value_bound = {'low': 1, 'up': 10}
kernel_coefficient_bound = {'low': -8, 'up': -1}
problem = SVC_2d.SVC_2D(x, y, regularization_value_bound, kernel_coefficient_bound)
method_params = SolverParameters(r=np.double(2.0), itersLimit=200)
method_params = SolverParameters(r=np.double(2.0), iters_limit=200)
solver = Solver(problem, parameters=method_params)
apl = AnimationNDPaintListener(varsIndxs=[0, 1], toPaintObjFunc=False)
apl = AnimationNDPaintListener(vars_indxs=[0, 1], to_paint_obj_func=False)
solver.AddListener(apl)
spl = StaticNDPaintListener(varsIndxs=[0, 1], mode="surface", calc="interpolation")
spl = StaticNDPaintListener(vars_indxs=[0, 1], mode="surface", calc="interpolation")
solver.AddListener(spl)
cfol = ConsoleFullOutputListener(mode='full')
solver.AddListener(cfol)
Expand Down
Loading