from sklearn import preprocessing

import pandas as pd

import numpy as np

from xgboost import XGBRegressor

import xgboost as xgb

from sklearn.model_selection import train_test_split

from sklearn.metrics import accuracy_score

from sklearn.feature_extraction import DictVectorizer

from sklearn.linear_model import LinearRegression

from sklearn.model_selection import GridSearchCV

df = pd.read_csv('t_car.csv')

df = df.dropna(axis=0, how='any') # 删除空值数据 axis=0行，axis=1列

df = df.drop(['title'], axis=1)

print(df.info())

a = ['gear_type'] # 需要OneHotEncoder的feature

le = preprocessing.LabelEncoder()

le_count = 0

label_code_map = {} # 保存编码的数据，在预测时需要将数据根据训练数据编码

for col in df:

if df[col].dtype == 'object':

# If 2 or fewer unique categories

if len(list(df[col].unique())) >= 10:

# Train on the training data

le.fit(df[col])

# Transform both training and testing data

orl = df[col]

df[col] = le.transform(orl)

label_code_map[col] = le.classes_

# Keep track of how many columns were label encoded

le_count += 1

print('%d columns were label encoded.' % le_count)

datax = pd.get_dummies(df[a]) # one_hot

df = df.join(datax)

df = df.drop(a, axis=1)

y_train = df.price

x_train = df.drop(['price'], axis=1) # 删除标签

x_train, x_test, y_train, y_test = train_test_split(x_train, y_train, test_size=0.3, shuffle=True)

model = XGBRegressor(max_depth=300, learning_rate=0.01, n_estimators=700,

missing=None, importance_type='gain')

model.fit(x_train, y_train)

cv_params = {'n_estimators': [500, 550, 600, 650, 700], 'max_depth': [3, 4, 5, 6, 7, 8, 9, 10], 'min_child_weight': [6, 7, 8],

'gamma': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6]}

optimized_GBM = GridSearchCV(estimator=model, param_grid=cv_params, scoring='r2', cv=3, verbose=1, n_jobs=-1)

optimized_GBM.fit(x_train, y_train)

evalute_result =optimized_GBM.return_train_score

print('每轮迭代运行结果:{0}'.format(evalute_result))

print('参数的最佳取值：{0}'.format(optimized_GBM.best_params_))

print('最佳模型得分:{0}'.format(optimized_GBM.best_score_))

print(len(le.classes_))

t = ['兰州', '雪佛兰', '2.4L', 2016, 8.9, 0, 20.61, 0, 1]

output_volume_l = label_code_map['output_volume']

city_l = label_code_map['city']

brand_l = label_code_map['brand']

def re_assign(index, ls, tbd):

break

re_assign(2, output_volume_l, t)

re_assign(0, city_l, t)

re_assign(1, brand_l, t)

tu = tuple(t)

t = list()

t.append(tu)

t = pd.DataFrame(t, columns=['city', 'brand', 'output_volume', 'launch_year', 'kilometres', 'is_import', 'old_price', 'gear_type_手动挡', 'gear_type_自动挡'])

y_ = model.predict(t)

print(y_)