Particle Swarm Optimization¶

Installation¶

In [1]:

!pip install zoofs

Requirement already satisfied: zoofs in c:\users\user\appdata\local\programs\python\python39\lib\site-packages (0.1.2)
Requirement already satisfied: plotly in c:\users\user\appdata\local\programs\python\python39\lib\site-packages (from zoofs) (5.3.1)
Requirement already satisfied: pandas in c:\users\user\appdata\local\programs\python\python39\lib\site-packages (from zoofs) (1.3.2)
Requirement already satisfied: numpy in c:\users\user\appdata\local\programs\python\python39\lib\site-packages (from zoofs) (1.21.2)
Requirement already satisfied: scipy in c:\users\user\appdata\local\programs\python\python39\lib\site-packages (from zoofs) (1.7.1)
Requirement already satisfied: pytz>=2017.3 in c:\users\user\appdata\local\programs\python\python39\lib\site-packages (from pandas->zoofs) (2021.1)
Requirement already satisfied: python-dateutil>=2.7.3 in c:\users\user\appdata\local\programs\python\python39\lib\site-packages (from pandas->zoofs) (2.8.2)
Requirement already satisfied: six>=1.5 in c:\users\user\appdata\local\programs\python\python39\lib\site-packages (from python-dateutil>=2.7.3->pandas->zoofs) (1.16.0)
Requirement already satisfied: tenacity>=6.2.0 in c:\users\user\appdata\local\programs\python\python39\lib\site-packages (from plotly->zoofs) (8.0.1)

WARNING: You are using pip version 21.1.1; however, version 21.2.4 is available.
You should consider upgrading via the 'c:\users\user\appdata\local\programs\python\python39\python.exe -m pip install --upgrade pip' command.

Load Breast cancer dataset¶

In [2]:

from sklearn.datasets import load_breast_cancer
import pandas as pd
data = load_breast_cancer()
X_train=pd.DataFrame(data['data'],columns=data['feature_names'])
y_train=pd.Series(data['target'])

Importing Zoofs Algo¶

In [3]:

from zoofs import ParticleSwarmOptimization

Setting up Objective function and fitting algo¶

In [4]:

from sklearn.metrics import log_loss
# define your own objective function, make sure the function receives four parameters,
#  fit your model and return the objective value ! 
def objective_function_topass(model,X_train, y_train, X_valid, y_valid):      
    model.fit(X_train,y_train)  
    P=log_loss(y_valid,model.predict_proba(X_valid))
    return P
    
# create object of algorithm
algo_object=ParticleSwarmOptimization(objective_function_topass,n_iteration=25,
                                       population_size=20,minimize=True)
import lightgbm as lgb
lgb_model = lgb.LGBMClassifier()                                       
# fit the algorithm
algo_object.fit(lgb_model,X_train, y_train, X_train, y_train,verbose=True)
#plot your results

		 Best value of metric across iteration 	 Best value of metric across population  
Iteration 0 	 0.0005296967375056357 					 0.0005296967375056357 
Iteration 1 	 0.0005350875802894996 					 0.0005296967375056357 
Iteration 2 	 0.0004900565872079092 					 0.0004900565872079092 
Iteration 3 	 0.00048332067099894437 					 0.00048332067099894437 
Iteration 4 	 0.00045637897622823905 					 0.00045637897622823905 
Iteration 5 	 0.000466775244080281 					 0.00045637897622823905 
Iteration 6 	 0.0004918998489345662 					 0.00045637897622823905 
Iteration 7 	 0.0004921178797640744 					 0.00045637897622823905 
Iteration 8 	 0.00042340845980268525 					 0.00042340845980268525 
Iteration 9 	 0.00042474680152597216 					 0.00042340845980268525 
Iteration 10 	 0.0003983627188019335 					 0.0003983627188019335 
Iteration 11 	 0.0004249661758924971 					 0.0003983627188019335 
Iteration 12 	 0.0004264664684658149 					 0.0003983627188019335 
Iteration 13 	 0.00046151713507627454 					 0.0003983627188019335 
Iteration 14 	 0.0004123962375142004 					 0.0003983627188019335 
Iteration 15 	 0.00043259241757591187 					 0.0003983627188019335 
Iteration 16 	 0.00039191564861833434 					 0.00039191564861833434 
Iteration 17 	 0.0004517608567341095 					 0.00039191564861833434 
Iteration 18 	 0.00040691262960006647 					 0.00039191564861833434 
Iteration 19 	 0.0004248262025278297 					 0.00039191564861833434 
Iteration 20 	 0.0004102547126767602 					 0.00039191564861833434 
Iteration 21 	 0.00041051318455042024 					 0.00039191564861833434 
Iteration 22 	 0.00037769760489395467 					 0.00037769760489395467 
Iteration 23 	 0.0004249105077075491 					 0.00037769760489395467 
Iteration 24 	 0.000432562624556134 					 0.00037769760489395467 

Out[4]:

['mean perimeter',
 'mean area',
 'mean smoothness',
 'mean concavity',
 'mean concave points',
 'mean fractal dimension',
 'radius error',
 'texture error',
 'perimeter error',
 'area error',
 'smoothness error',
 'compactness error',
 'concavity error',
 'concave points error',
 'symmetry error',
 'fractal dimension error',
 'worst radius',
 'worst texture',
 'worst perimeter',
 'worst smoothness',
 'worst compactness',
 'worst symmetry',
 'worst fractal dimension']

Plotting the Results¶

In [5]:

algo_object.plot_history()