#! /usr/bin/env python3
#
def diabetes ( ):

#*****************************************************************************80
#
## diabetes() does logistic regression on diabetes diagnosis data.
#
#  Licensing:
#
#    This code is distributed under the GNU LGPL license.
#
#  Modified:
#
#    14 February 2022
#
#  Author:
#
#    John Burkardt
#
  from sklearn.linear_model import LogisticRegression
  import matplotlib.pyplot as plt
  import numpy as np

  print ( '' )
  print ( 'diabetes():' )
  print ( '  Logistic regression for diabetes diagnosis data.' )
#
#  Read the data:
#
  data = np.loadtxt ( 'diabetes_data.txt' )
  prg = data[:,0]
  glu = data[:,1]
  dbp = data[:,2]
  tri = data[:,3]
  ins = data[:,4]
  bmi = data[:,5]
  ped = data[:,6]
  age = data[:,7]
  dia = data[:,8]

  n = len ( prg )

  X = np.zeros ( [ n, 9 ] )
  X[:,0] = 1
  X[:,1] = prg[:]
  X[:,2] = glu[:]
  X[:,3] = dbp[:]
  X[:,4] = tri[:]
  X[:,5] = ins[:]
  X[:,6] = bmi[:]
  X[:,7] = ped[:]
  X[:,8] = age[:]

  y = dia[:]
#
#  Scale the data.
#
  from sklearn import preprocessing
  scaler = preprocessing.StandardScaler().fit ( X )
  X_scaled = scaler.transform ( X )
#
#  Fit the logistic regression model to the data.
#
  classifier = LogisticRegression(random_state=0).fit ( X_scaled, y )
#
#  Evaluate the model on the original data.
#
  yp = classifier.predict ( X_scaled )
#
#  Compute MSE
#
  mse = ( 1.0 / n ) * sum ( ( yp - y )**2 )
  print ( '  MSE = ', mse )
#
#  Construct the coefficient array W.
#
  w = classifier.coef_[0].copy()
  w[0] = classifier.intercept_

  print ( '' )
  print ( '  Computed weights W = ', w )
#
#  Evaluate the admissions formula for new data.
#
#  Raw values:
#    1  46  81
#    2  26  75
#    3  30  66
#
  return

  X2 = np.array ( [ \
    [ 1.0, 46, 81 ], \
    [ 1.0, 26, 75 ], \
    [ 1.0, 30, 66 ] ] )
  print ( '  New medical data:' )
  print ( X2 )
  y2 = classifier.predict ( X2 )
  print ( '  diabetes new student? (0/1)' )
  print ( y2 )
  p2 = classifier.predict_proba ( X2 )
  print ( '  Strength of decision (0/1)' )
  print ( p2 )
#
#  Terminate.
#
  print ( '' )
  print ( 'diabetes():' )
  print ( '  Normal end of execution.' )

  return

if ( __name__ == '__main__' ):
  diabetes ( )