Tue May 20 22:22:19 2025

perceptron_test():
  python version: 3.10.12
  numpy version:  1.26.4
  Use the perceptron algorithm to determine a classifier
  for sets of data.

dividing_line_test():
  Apply the perceptron to an artificial data set.
  Points (x,y) are in set 0 if y < 3x + 2
  Points (x,y) are in set 1 if y > 3x + 2
  Generate 25 points at random and search for a classifier
  using the perceptron algorithm

  Number of training data values =  25

  Training data:
[[0.00000000e+00 5.48853558e-01 2.28520448e+00 0.00000000e+00]
 [1.00000000e+00 2.09172037e-01 2.46961184e+00 0.00000000e+00]
 [2.00000000e+00 2.37940914e-01 2.91613866e+00 1.00000000e+00]
 [3.00000000e+00 2.63612220e-01 4.98364220e+00 1.00000000e+00]
 [4.00000000e+00 1.74363724e-01 2.55737812e+00 1.00000000e+00]
 [5.00000000e+00 6.44138098e-01 4.86159269e+00 1.00000000e+00]
 [6.00000000e+00 2.68704314e-01 4.17198630e+00 1.00000000e+00]
 [7.00000000e+00 6.45885709e-01 2.29719923e+00 0.00000000e+00]
 [8.00000000e+00 4.80114400e-01 4.33877905e+00 1.00000000e+00]
 [9.00000000e+00 3.58845448e-01 4.41496823e+00 1.00000000e+00]
 [1.00000000e+01 5.07678780e-02 2.09820920e+00 0.00000000e+00]
 [1.10000000e+01 4.02045786e-01 3.42603110e+00 1.00000000e+00]
 [1.20000000e+01 8.57537429e-01 2.26282105e+00 0.00000000e+00]
 [1.30000000e+01 6.12238953e-01 2.38061752e+00 0.00000000e+00]
 [1.40000000e+01 3.33987334e-02 2.90741765e+00 1.00000000e+00]
 [1.50000000e+01 8.42187155e-01 2.91825224e+00 0.00000000e+00]
 [1.60000000e+01 7.20420338e-01 3.39020872e+00 0.00000000e+00]
 [1.70000000e+01 4.97258744e-01 3.90484925e+00 1.00000000e+00]
 [1.80000000e+01 3.36950334e-01 2.45559661e+00 0.00000000e+00]
 [1.90000000e+01 8.93034401e-03 4.62364764e+00 1.00000000e+00]
 [2.00000000e+01 7.67367027e-01 2.70743383e+00 0.00000000e+00]
 [2.10000000e+01 1.38427468e-01 3.18510793e+00 1.00000000e+00]
 [2.20000000e+01 3.15829884e-01 2.96731111e+00 1.00000000e+00]
 [2.30000000e+01 7.55613686e-02 2.61370788e+00 1.00000000e+00]
 [2.40000000e+01 9.14990441e-02 4.44844112e+00 1.00000000e+00]]
  25 good y, and 25 bad y
  0.00893034 <= X <= 0.857537
  2.09821 <= Y <= 4.98364
  GOOD: mean X = 0.266045, mean Y = 3.75473
  BAD:  mean X = 0.559138, mean Y = 2.52652
  Graphics saved as "dividing_line_data.png"
  Using learning rate alpha = 0.05
  Using  1000 steps
  Iteration terminated without convergence on step 1000
  e = 6

  Perceptron weights:
  f(x) = -0.771604 + -0.472154 * x + 0.42626 * y

  Index    x*w  (0<x*w)  Rating

      0   -0.06      0      0
      1    0.18      1      0
      2    0.36      1      1
      3    1.23      1      1
      4    0.24      1      1
      5    1.00      1      1
      6    0.88      1      1
      7   -0.10      0      0
      8    0.85      1      1
      9    0.94      1      1
     10    0.10      1      0
     11    0.50      1      1
     12   -0.21      0      0
     13   -0.05      0      0
     14    0.45      1      1
     15    0.07      1      0
     16    0.33      1      0
     17    0.66      1      1
     18    0.12      1      0
     19    1.20      1      1
     20    0.02      1      0
     21    0.52      1      1
     22    0.34      1      1
     23    0.31      1      1
     24    1.08      1      1
  Graphics saved as "dividing_line_classified.png"

generator_ratings_test():
  Apply the perceptron to generator ratings data.
  Data file is " generators.txt "
  Data is index, rpm, vibration, rating
  Rating is 0 for bad, 1 for good.
  Use the perceptron algorithm to determine a
  classifier f(rpm,vibration) -> {0, 1}

  Generator ratings

  Number of generators = 56
  56 good generators, and 56 bad generators
  562 <= RPM <= 939
  79 <= VIB <= 585
  GOOD: mean RPM = 740.75, mean VIB = 447.286
  BAD:  mean RPM = 741.964, mean VIB = 219.857
  Graphics saved as "generator_ratings_data.png"
  All training data classified on step 8

  Perceptron weights:
  f(x) = -0.541674 + 0.397848 * r + 0.740477 * v


  Index    x*w  (0<x*w)  Rating

      0   -0.21      0      0
      1    0.07      1      1
      2    0.07      1      1
      3    0.06      1      1
      4    0.12      1      1
      5   -0.01      0      0
      6   -0.29      0      0
      7   -0.23      0      0
      8    0.10      1      1
      9   -0.02      0      0
     10   -0.05      0      0
     11    0.21      1      1
     12   -0.11      0      0
     13    0.18      1      1
     14   -0.21      0      0
     15   -0.23      0      0
     16    0.17      1      1
     17    0.23      1      1
     18   -0.22      0      0
     19    0.24      1      1
     20    0.23      1      1
     21    0.20      1      1
     22   -0.11      0      0
     23   -0.21      0      0
     24    0.14      1      1
     25    0.12      1      1
     26   -0.10      0      0
     27   -0.09      0      0
     28    0.12      1      1
     29   -0.06      0      0
     30    0.20      1      1
     31   -0.17      0      0
     32   -0.11      0      0
     33    0.20      1      1
     34   -0.08      0      0
     35   -0.23      0      0
     36    0.24      1      1
     37    0.29      1      1
     38   -0.12      0      0
     39   -0.21      0      0
     40    0.30      1      1
     41   -0.08      0      0
     42   -0.13      0      0
     43   -0.16      0      0
     44    0.10      1      1
     45    0.21      1      1
     46   -0.21      0      0
     47   -0.10      0      0
     48    0.26      1      1
     49    0.22      1      1
     50    0.17      1      1
     51    0.28      1      1
     52    0.28      1      1
     53   -0.14      0      0
     54    0.19      1      1
     55   -0.18      0      0
  Graphics saved as "generator_ratings_classified.png"

perceptron_test():
  Normal end of execution.
Tue May 20 22:22:20 2025