pythonで機械学習するときの自分用テンプレート

基本テンプレート

import numpy as np
import pandas as pd

dataset = pd.read_csv("CSVファイル名")
def ping(data):
    df = data.dropna()
    df = pd.get_dummies(df, drop_first=True)
    return df

train = ping(dataset)

from sklearn.model_selection import train_test_split
x = train.drop("答えの名", axis=1).values
t = train["答えの名"]

x_train_val, x_test, t_train_val, t_test = train_test_split(x, t, test_size=0.2, random_state=1)
x_train, x_val, t_train, t_val = train_test_split(x_train_val, t_train_val, test_size=0.3, random_state=1)

from sklearn.tree import DecisionTreeClassifier
from sklearn.model_selection import RandomizedSearchCV

estimator = DecisionTreeClassifier(random_state=0)

param_distributions = {
    'max_depth': list(range(5, 100, 2)),#開始値,終了値,ステップ（刻み値）
    'min_samples_split': list(range(2, 50, 1))
}

#試行回数
n_iter = 100

cv = 5

tuned_model = RandomizedSearchCV(
    estimator=estimator, 
    param_distributions=param_distributions, 
    n_iter=n_iter, cv=cv, 
    random_state=0, return_train_score=False
)

tuned_model.fit(x_train_val, t_train_val)

dd = pd.DataFrame(tuned_model.cv_results_).sort_values('rank_test_score').T
dd.to_csv("me.csv")

best_model = tuned_model.best_estimator_

print(best_model.score(x_train_val, t_train_val))
print(best_model.score(x_test, t_test))

予測

#テストデータ
test_data = [[0,0],[1,1]]
test_label = best_model.predict(test_data)
print("予測結果→")
print(test_label)