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

#*****************************************************************************80
#
## admit does logistic regression on college admission data.
#
#  Licensing:
#
#    This code is distributed under the GNU LGPL license.
#
#  Modified:
#
#    13 October 2019
#
#  Author:
#
#    John Burkardt
#
  import matplotlib.pyplot as plt
  import numpy as np
  from logistic_regression import logistic_regression

  print ( '' )
  print ( 'admit:' )
  print ( '  We have M=100 english and math test scores' )
  print ( '  and the corresponding college admission decision y.' )
  print ( '  Consider the model' )
  print ( '    y(x) = 1 / ( 1 + exp ( - w\'x ) )' )
  print ( '  Determine weights w, and apply to 3 new cases.' )
#
#  Read the data: (x1=english, x2=math, y=admit(0/1))
#
  data = np.loadtxt ( 'admit_data.txt' )
  eng = data[:,0]
  mat = data[:,1]
  admit = data[:,2]
  m = len ( eng )
#
#  Normalize the input data.
#
  eng = ( eng - np.min ( eng ) ) / ( np.max ( eng ) - np.min ( eng ) )
  mat = ( mat - np.min ( mat ) ) / ( np.max ( mat ) - np.min ( mat ) )
#
#  Create arrays x and y.
#
  x = np.zeros ( [ m, 3 ] )
  x[:,0] = 1
  x[:,1] = eng[:]
  x[:,2] = mat[:]

  y = admit[:]

  alpha = 0.01
  kmax = 10000

  w = logistic_regression ( x, y, alpha, kmax )

  print ( '' )
  print ( '  Using ALPHA = %g' % ( alpha ) )
  print ( '  Final W = (%g,%g,%g)' % ( w[0], w[1], w[2] ) )
#
#  Display data.
#
  plt.plot ( x[y==0,1], x[y==0,2], 'r.', markersize = 15 )
  plt.plot ( x[y==1,1], x[y==1,2], 'b.', markersize = 15 )
  plt.plot ( [-w[0]/w[1], (-w[0]-w[2])/w[1]], [0,1], 'k-', linewidth = 3 )
  plt.xlabel ( '<-- English Exam -->' )
  plt.ylabel ( '<-- Math Exam -->' )
  plt.title ( 'Blue = Admitted, Red = Not admitted', fontsize = 16 )
  plt.grid ( True )
  filename = 'admit.png'
  plt.savefig ( filename )
  plt.show ( )
  print ( '' )
  print ( '  Graphics saved as "%s"' % ( filename ) )
#
#  Evaluate the admissions formula for new data.
#
#  Raw values:
#    1  46  81
#    2  26  75
#    3  30  66
#
#  Normalized values:
#
#    1  0.66  0.88
#    2  0.22  0.74
#    3  0.31  0.55
#
  x1 = np.array ( [ 1.0, 0.66, 0.88 ] )
  wtx = np.dot ( w, x1 )
  y1 = 1.0 / ( 1.0 + np.exp ( - wtx ) )
  print ( 'Student 1: wtx = ', wtx, '  Y = ', y1 )

  x2 = np.array ( [ 1.0, 0.22, 0.74 ] )
  wtx = np.dot ( w, x2 )
  y2 = 1.0 / ( 1.0 + np.exp ( - wtx ) )
  print ( 'Student 2: wtx = ', wtx, '  Y = ', y2 )

  x3 = np.array ( [ 1.0, 0.31, 0.55 ] )
  wtx = np.dot ( w, x3 )
  y3 = 1.0 / ( 1.0 + np.exp ( - wtx ) )
  print ( 'Student 3: wtx = ', wtx, '  Y = ', y3 )
#
#  Terminate.
#
  print ( '' )
  print ( 'admit:' )
  print ( '  Normal end of execution.' )

  return

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