import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.datasets import load_wine
 
wine = load_wine()
data = wine.data  # 數(shù)據(jù)
lables = wine.target  # 標簽
feaures = wine.feature_names
df = pd.DataFrame(data, columns=feaures)  # 原始數(shù)據(jù)
 
# 第一步：無量綱化
def standareData(df):
    """
    df : 原始數(shù)據(jù)
    return : data 標準化的數(shù)據(jù)
    """
    data = pd.DataFrame(index=df.index)  # 列名，一個新的dataframe
    columns = df.columns.tolist()  # 將列名提取出來
    for col in columns:
        d = df[col]
        max = d.max()
        min = d.min()
        mean = d.mean()
        data[col] = ((d - mean) / (max - min)).tolist()
    return data
 
#  某一列當做參照序列，其他為對比序列
def graOne(Data, m=0):
    """
    return:
    """
    columns = Data.columns.tolist()  # 將列名提取出來
    # 第一步：無量綱化
    data = standareData(Data)
    referenceSeq = data.iloc[:, m]  # 參考序列
    data.drop(columns[m], axis=1, inplace=True)  # 刪除參考列
    compareSeq = data.iloc[:, 0:]  # 對比序列
    row, col = compareSeq.shape
    # 第二步：參考序列 - 對比序列
    data_sub = np.zeros([row, col])
    for i in range(col):
        for j in range(row):
            data_sub[j, i] = abs(referenceSeq[j] - compareSeq.iloc[j, i])
    # 找出最大值和最小值
    maxVal = np.max(data_sub)
    minVal = np.min(data_sub)
    cisi = np.zeros([row, col])
    for i in range(row):
        for j in range(col):
            cisi[i, j] = (minVal + 0.5 * maxVal) / (data_sub[i, j] + 0.5 * maxVal)
    # 第三步：計算關聯(lián)度
    result = [np.mean(cisi[:, i]) for i in range(col)]
    result.insert(m, 1)  # 參照列為1
    return pd.DataFrame(result)
 
def GRA(Data):
    df = Data.copy()
    columns = [str(s) for s in df.columns if s not in [None]]  # [1 2 ,,,12]
    # print(columns)
    df_local = pd.DataFrame(columns=columns)
    df.columns = columns
    for i in range(len(df.columns)):  # 每一列都做參照序列，求關聯(lián)系數(shù)
        df_local.iloc[:, i] = graOne(df, m=i)[0]
    df_local.index = columns
    return df_local
 
# 熱力圖展示
def ShowGRAHeatMap(DataFrame):
    colormap = plt.cm.hsv
    ylabels = DataFrame.columns.values.tolist()
    f, ax = plt.subplots(figsize=(15, 15))
    ax.set_title('Wine GRA')
    # 設置展示一半，如果不需要注釋掉mask即可
    mask = np.zeros_like(DataFrame)
    mask[np.triu_indices_from(mask)] = True  # np.triu_indices 上三角矩陣
 
    with sns.axes_style("white"):
        sns.heatmap(DataFrame,
                    cmap="YlGnBu",
                    annot=True,
                    mask=mask,
                    )
    plt.show()
 
data_wine_gra = GRA(df)
ShowGRAHeatMap(data_wine_gra)

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
 
# ====熱力圖
from matplotlib.ticker import FormatStrFormatter
encoding="utf-8"
data = pd.read_csv("tu.csv", encoding="utf-8")     #讀取數(shù)據(jù)
data.drop_duplicates()
data.columns = [i for i in range(data.shape[1])]
 
# 計算兩兩屬性之間的皮爾森相關系數(shù)
corrmat = data.corr()
f, ax = plt.subplots(figsize=(12, 9))
 
# 返回按“列”降序排列的前n行
k = 30
cols = corrmat.nlargest(k, data.columns[0]).index
 
# 返回皮爾遜積矩相關系數(shù)
cm = np.corrcoef(data[cols].values.T)
sns.set(font_scale=1.25)
hm = sns.heatmap(cm,
                 cbar=True,
                 annot=True,
                 square=True,
                 fmt=".3f",
                 vmin=0,             #刻度閾值
                 vmax=1,
                 linewidths=.5,
                 cmap="RdPu",        #刻度顏色
                 annot_kws={"size": 10},
                 xticklabels=True,
                 yticklabels=True)             #seaborn.heatmap相關屬性
# 解決中文顯示問題
plt.rcParams['font.sans-serif'] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False
# plt.ylabel(fontsize=15,)
# plt.xlabel(fontsize=15)
plt.title("主要變量之間的相關性強弱", fontsize=20)
plt.show()

seaborn.heatmap(data, vmin=None, 
vmax=None, cmap=None, center=None, 
robust=False, annot=None, fmt='.2g', 
annot_kws=None, linewidths=0, linecolor='white',
 cbar=True, cbar_kws=None, cbar_ax=None, 
square=False, xticklabels='auto', yticklabels='auto',
 mask=None, ax=None, **kwargs)

sns.heatmap(
      data,
      vmin=None,
      vmax=None,
      cmap=None,
      center=None,
      robust=False,
      annot=None,
      fmt='.2g',
      annot_kws=None,
      linewidths=0,
      linecolor=‘white',
      cbar=True,
      cbar_kws=None,
      cbar_ax=None,
      square=False,
      xticklabels=‘auto',
      yticklabels=‘auto',
      mask=None,
      ax=None,
)