from statsmodels.stats.outliers_influence import variance_inflation_factor
from statsmodels.tools.tools import add_constant
import pandas as pd
from patsy import dmatrices
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
#https://www.codecamp.ru/blog/how-to-calculate-vif-in-python/   for each independent model
#https://future-chem.com/esol-reg-aic/
#https://stackoverflow.com/questions/42658379/variance-inflation-factor-in-python
#https://www.hse.ru/data/2010/10/22/1222676695/%D0%A7%D0%B0%D1%81%D1%82%D1%8C%204%20%D0%9C%D1%83%D0%BB%D1%8C%D1%82%D0%B8%D0%BA%D0%BE%D0%BB%D0%BB%D0%B8%D0%BD%D0%B5%D0%B0%D1%80%D0%BD%D0%BE%D1%81%D1%82%D1%8C..pdf
"""df = pd.DataFrame(
    {'a': [1, 1, 2, 3, 4],
     'b': [2, 2, 3, 2, 1],
     'c': [4, 6, 7, 8, 9],
     'd': [4, 3, 4, 5, 4]}
)"""
plt.figure(figsize=(6, 6))
file_name = "Test"
df = pd.read_excel(u"test_data", sheet_name="Sheet1")
df_new = df.dropna()
stop = 0
fig = plt.figure(figsize=(7, 7))
col_names = []

for x in df_new.columns:
    col_names.append(x)

#matrix = sns.heatmap(df_cor, annot=True)

#plt.savefig('heatmap_new_1.png', dpi=300, bbox_inches='tight')

#ds=(np.linalg.inv(df_cor))
df_cor = df_new.corr()
dfs = pd.DataFrame(np.linalg.inv(df_cor), df_cor.columns, df_cor.index)

for x in col_names:
    max_int = max(np.diag(dfs))
    if dfs[f'{x}'] == max(np.diag(dfs)) & dfs[f'{x}'] >= 10)]:
        df_new.drop(dfs[f'{x}'] == max(np.diag(dfs)))
        df_cor = df_new.corr()
        dfs = pd.DataFrame(np.linalg.inv(df_cor), df_cor.columns, df_cor.index)

#matrix_1 = sns.heatmap(dfs, annot=True)

#plt.savefig('heatmap_5.png', dpi=300, bbox_inches='tight')

print(df)
"""df = pd.DataFrame({'rating': [90, 85, 82, 88, 94, 90, 76, 75, 87, 86],
 'points': [25, 20, 14, 16, 27, 20, 12, 15, 14, 19],
 'assists': [5, 7, 7, 8, 5, 7, 6, 9, 9, 5],
 'rebounds': [11, 8, 10, 6, 6, 9, 6, 10, 10, 7]})
"""
X = add_constant(df)
ds=(np.linalg.inv(np.corrcoef(df, rowvar=0)))
print(ds)
#Источник: https://pythonpip.ru/examples/obnaruzhenie-multikollinearnosti-vif-v-python
#print (X)

y, D = dmatrices('b ~ a+c+b', data=df, return_type='dataframe')
#calculate VIF for each explanatory variable
vif = pd.DataFrame()
vif['VIF'] = [variance_inflation_factor(D.values, i) for i in range(D.shape[1])]
vif['variable'] = D.columns
vif

stop = 0