PyGAD传递整数参数不符合文档要求的问题排查与解决

问题背景 #

使用PyGAD优化随机森林模型超参数时，gene_space中通过np.linspace生成的整数数组，经PyGAD处理后传递给RandomForestClassifier时被转为numpy.float64类型，触发报错：

ValueError: n_estimators must be an integer, got <class 'numpy.float64'>

直接调用gene_space[0][0]作为参数可正常运行，确认问题源于PyGAD对基因值的类型处理。

相关代码 #

gene_space定义

import numpy as np
gene_space = [ 
    # n_estimators
    np.linspace(50,200,25, dtype='int'),
    # min_samples_split
    np.linspace(2,10,5, dtype='int'),
    # min_samples_leaf
    np.linspace(1,10,5, dtype='int'),
    # min_impurity_decrease
    np.linspace(0,1,10, dtype='float')
]

适应度函数

from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.utils import sample_without_replacement

def fitness_function_factory(data=data, y_name='y', sample_size=100):
    def fitness_function(solution, solution_idx):
        model = RandomForestClassifier(
            n_estimators=solution[0],
            min_samples_split=solution[1],
            min_samples_leaf=solution[2],
            min_impurity_decrease=solution[3]
        )
        
        X = data.drop(columns=[y_name])
        y = data[y_name]
        X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5)
        
        train_idx = sample_without_replacement(n_population=len(X_train), 
                                              n_samples=sample_size)
        test_idx = sample_without_replacement(n_population=len(X_test), 
                                              n_samples=sample_size)
         
        model.fit(X_train.iloc[train_idx], y_train.iloc[train_idx])
        fitness = model.score(X_test.iloc[test_idx], y_test.iloc[test_idx])
        
        return fitness 
    return fitness_function

PyGAD实例化

import pygad

cross_validate = pygad.GA(gene_space=gene_space,
                      fitness_func=fitness_function_factory(),
                      num_generations=100,
                      num_parents_mating=2,
                      sol_per_pop=8,
                      num_genes=len(gene_space),
                      parent_selection_type='sss',
                      keep_parents=2,
                      crossover_type="single_point",
                      mutation_type="random",
                      mutation_percent_genes=25)

cross_validate.best_solution()

解决方法 #

1. 适应度函数内强制类型转换 #

在创建RandomForestClassifier时，将需要整数的参数显式转为int类型，直接规避PyGAD的类型转换问题：

def fitness_function(solution, solution_idx):
    model = RandomForestClassifier(
        n_estimators=int(solution[0]),
        min_samples_split=int(solution[1]),
        min_samples_leaf=int(solution[2]),
        min_impurity_decrease=solution[3]
    )
    # 后续代码不变

2. 将numpy数组转为Python原生列表 #

PyGAD对原生Python整数的处理更稳定，把gene_space中的numpy数组转为列表：

gene_space = [ 
    np.linspace(50,200,25, dtype='int').tolist(),
    np.linspace(2,10,5, dtype='int').tolist(),
    np.linspace(1,10,5, dtype='int').tolist(),
    np.linspace(0,1,10, dtype='float').tolist()
]

3. 统一处理参数类型 #

在适应度函数开头对所有参数做类型映射，让代码更整洁：

def fitness_function(solution, solution_idx):
    # 前3个参数转整数，第4个保留浮点
    processed_sol = [int(x) for x in solution[:3]] + [solution[3]]
    model = RandomForestClassifier(
        n_estimators=processed_sol[0],
        min_samples_split=processed_sol[1],
        min_samples_leaf=processed_sol[2],
        min_impurity_decrease=processed_sol[3]
    )
    # 后续代码不变

内容的提问来源于stack exchange，提问作者jbuddy_13