deqin - 预测鸢尾花

# 1.提取数据
# data 数据 set集合   path 数据
from sklearn import datasets

flower_data = datasets.load_iris()
print(flower_data)
# 花萼的长度  花萼的宽度   花瓣的长度   花瓣的宽度
# 'setosa'鸢尾花, 'versicolor'变色鸢尾花, 'virginica'维吉尼亚鸢尾花
#  物种分类 科学家
# 表格  target 目标
import pandas as pd

hua = pd.DataFrame(flower_data.data)
species = pd.DataFrame(flower_data.target)
# 合并
df = pd.concat([hua, species], axis=1)
df.columns = ["花萼长度", "花萼宽度", "花瓣长度", "花瓣宽度", "种类"]

# 2.画图，能不能区别开
import matplotlib.pyplot as plt

# a = 1
# for i in ["花萼长度", "花萼宽度", "花瓣长度", "花瓣宽度"]:
#     ax = plt.subplot(1,4,a)
#     a = a+1
#     plt.rcParams["font.sans-serif"] = ["SimHei"]
#     ax.scatter(df["花萼长度"][0:49], df[i][0:49], color="blue", label="鸢尾花")
#     ax.scatter(df["花萼长度"][50:99], df[i][50:99], color="green", label="变色鸢尾花")
#     ax.scatter(df["花萼长度"][100:149], df[i][100:149], color="red", label="维吉尼亚鸢尾花")
#     plt.legend() # 图例
#     plt.xlabel("花萼长度")
#     plt.ylabel("%s"%i)
#     plt.title("花萼长度和%s的关系"%i)
#     plt.show()
# 3.预测
import numpy as np
deqinhua = np.asarray([[10, 20, 19, 34]])
from sklearn import linear_model
xun_lian_qi = linear_model.LogisticRegression()
xun_lian_qi.fit(hua,species)
score = xun_lian_qi.score(hua,species)
print(score)
print(xun_lian_qi.predict(deqinhua))