{مصطفى} \ {mustafa}

# Select algorithms based on how they function

from mustafapy import mustafa
m = mustafa(atype='class') // "reg" for regression
m.hypers()

# Fetches all available classification models and their parameters

# Select algorithms by their names

from mustafapy import mustafa
m = mustafa(selection=['RadiusNeighborsClassifier', 'BernoulliNB', 'CategoricalNB'])
m.hypers()

# Fetches the listed and their parameters

# The default value for 'precision' is 1, the higher the value the more variety in numerical
# Hyperstep is the step between the generated numbers
# hyperparameters values

from mustafapy import mustafa
m = mustafa(precision=3,hyperstep=3)
m.hypers()

# every numerical hyperparameter will have at most 3 values set

# Set the number of hyperparameters to fetch for each algorith, default value is "max" which
# fetches all of them

from mustafapy import mustafa
m = mustafa(maxattrs=1)
m.hypers()

# for each algorithm only the first two hyperparameters will be fetched

# Assuming your cleaned dataset is in X and y variables


from mustafapy import mustafa
m = mustafa(selection=['MLPClassifier', 'KNeighborsClassifier', 'NearestCentroid'],
                    maxattrs=2,
                    precision=3)
m.gridsearch(X,y)

# 'MLPClassifier', 'KNeighborsClassifier', 'NearestCentroid' these algorithms with 2 attributes and 3
# values at most for each attribute will be fetched,