sk-learn Facebook���ݼ�Ԥ��ǩ��λ��

def facebook_demo():
    """
    sk-learn Facebook���ݼ�Ԥ��ǩ��λ��
    :return:
    """
    # 1����ȡ���ݼ�
    facebook = pd.read_csv('/Users/maxinze/Downloads/����ѧxiday2����/02-����/FBlocation/train.csv')

    # 2.�������ݴ���
    # 2.1 ��С���ݷ�Χ
    # ѡ��（2,2.5）��һ��Χ������，ʹ�� query
    facebook_data = facebook.query("x>5.0 & x<6 & y>5.0 & y<6.0")

    # 2.2 ѡ��ʱ������
    # ��ȡʱ��
    time = pd.to_datetime(facebook_data["time"], unit="s")
    time = pd.DatetimeIndex(time)
    # ��һ��day
    facebook_data["day"] = time.day
    # ��һ��hour
    facebook_data["hour"] = time.hour
    # ��һ��weekday
    facebook_data["weekday"] = time.weekday

    # 2.3 ȥ��ǩ�����ٵĵط�
    # �������，����Ŀ����
    place_count = facebook_data.groupby("place_id").count()
    # ѡ��ǩ������3��
    place_count = place_count[place_count["row_id"] > 3]
    # ��������
    facebook_data = facebook_data[facebook_data["place_id"].isin(place_count.index)]
    # facebook_data.shape()

    # 2.4 ɸѡ����ֵ��Ŀ��ֵ
    # ����ֵ
    x = facebook_data[["x", "y", "accuracy", "day", "hour", "weekday"]]
    # Ŀ��ֵ
    y = facebook_data["place_id"]

    # 2.5 �ָ����ݼ�（���ݼ�����） ��������ֵ， Ŀ��ֵ
    x_train, x_test, y_train, y_test = train_test_split(x, y, random_state=22)

    # 3.��������--����Ԥ����(��׼��)
    # 3.1 ʵ����һ��ת����
    transfer = StandardScaler()
    # 3.2 ����fit_transform
    # ����ѵ����
    x_train = transfer.fit_transform(x_train)
    # �������Լ�
    x_test = transfer.fit_transform(x_test)

    # 4.����ѧϰ--knn+cv
    # 4.1 ʵ����һ��������
    estimator = KNeighborsClassifier()
    # 4.2 ����gridsearchCV
    # param_grid = {"n_neighbors": [1, 3, 5, 7, 9]}
    param_grid = {"n_neighbors": [5, 7, 9]}
    estimator = GridSearchCV(estimator, param_grid=param_grid, cv=3 )
    # 4.3 ģ��ѵ��
    estimator.fit(x_train, y_train)

    # 5.ģ������
    # 5.1 ����������ʽ
    score = estimator.score(x_test, y_test)
    print("���Ԥ���׼ȷ��Ϊ:\n", score)

    y_predict = estimator.predict(x_test)
    print("����Ԥ��ֵΪ:\n", y_predict)
    print("Ԥ��ֵ����ʵֵ�ĶԱ����:\n", y_predict == y_test)

    # 5.2 ʹ�ý�����֤���������ʽ
    print("�ڽ�����֤����֤����ý��:\n", estimator.best_score_)
    print("��õIJ���ģ��:\n", estimator.best_estimator_)
    print("ÿ�ν�����֤�����֤��׼ȷ�ʽ����ѵ����׼ȷ�ʽ��:\n", estimator.cv_results_)

    return None

��Ϊֻѡ���˲��������ܴ���，����ģ��ѵ������Ե�׼ȷ�ʲ�̫��，�������ѡ��ȫ�������ܡ�