fork download
copy 
  1. import numpy as np
  2.  
  3.  
  4. def mean_normalization(Dataset):
  5.     max_first = np.max(Dataset[:, 1])
  6.     max_second = np.max(Dataset[:, 2])
  7.     min_first = np.min(Dataset[:, 1])
  8.     min_second = np.min(Dataset[:, 2])
  9.     mean_first = np.mean(Dataset[:, 1])
  10.     mean_second = np.mean(Dataset[:, 2])
  11.     for data in Dataset:
  12.         data[1] = (data[1] - mean_first) / (max_first - min_first)
  13.         data[2] = (data[2] - mean_second) / (max_second - min_second)
  14.     return Dataset, [max_first, min_first, mean_first, max_second, min_second, mean_second]
  15.  
  16.  
  17. def sigmoid(x):
  18.     return 1 / (1 + np.exp(-x))
  19.  
  20.  
  21. # Call for each training set, Individually.
  22.  
  23. def compute_cost(X, y, theta):
  24.     m = len(y)
  25.     h = sigmoid(np.dot(X, theta))
  26.     cost = (1 / m) * (np.dot(-y.T, np.log(h)) - np.dot((1 - y).T, np.log(1 - h)))
  27.     return cost
  28.  
  29.  
  30. # Call for each training set, Individually.
  31.  
  32.  
  33. def gradient_descent(X, y, theta, learning_rate, num_iterations):
  34.     m = len(y)
  35.     cost_history = []
  36.     for i in range(num_iterations):
  37.         theta = theta - (learning_rate / m) * (X.T @ (sigmoid(X @ theta) - y))
  38.         cost_history.append([compute_cost(X, y, theta)])
  39.     return cost_history, theta
  40.  
  41.  
  42. def prediction(data, optimum_theta):
  43.     data[1] = (data[1] - max_min_list[2]) / (max_min_list[0] - max_min_list[1])
  44.     data[2] = (data[2] - max_min_list[5]) / (max_min_list[3] - max_min_list[4])
  45.     sigmoid_value = sigmoid(data @ optimum_theta)
  46.     if sigmoid_value >= 0.5:
  47.         print("Student will be admitted.")
  48.     else:
  49.         print("Student will be rejected.")
  50.  
  51.  
  52. def initial_data():
  53.     dataset = np.loadtxt("Datasets/data1.txt", dtype=str)
  54.     Dataset = []
  55.     result = []
  56.  
  57.     for data in dataset:
  58.         data = data.split(",")
  59.         Dataset.append([1, float(data[0]), float(data[1])])
  60.         result.append([float(data[2])])
  61.     Dataset = np.array(Dataset, dtype=np.float64)
  62.     result = np.array(result, dtype=np.float64)
  63.     Dataset, max_min_list = mean_normalization(Dataset)
  64.     return Dataset, result, max_min_list
  65.  
  66.  
  67. def accuracy(X, y, theta, cutoff):
  68.     pred = [sigmoid(np.dot(X, theta)) >= cutoff]
  69.     acc = np.mean(pred == y)
  70.     print(acc * 100)
  71.  
  72.  
  73. X, y, max_min_list = initial_data()
  74. m, n = X.shape
  75. learning_rate = 0.1
  76. num_iterations = 400
  77. theta = np.zeros((n, 1))
  78. cost_history, optimum_theta = gradient_descent(X, y, theta, learning_rate, num_iterations)
  79. predict = np.array([1, 82.3, 93.3])
  80. prediction(predict, optimum_theta)
  81.  
  82.  
Runtime error #stdin #stdout #stderr 0.29s 36928KB
comments (?)
stdin
copy 
Standard input is empty
stdout
copy 
Standard output is empty
stderr
copy 
Traceback (most recent call last):
  File "./prog.py", line 73, in <module>
  File "./prog.py", line 53, in initial_data
  File "/usr/local/lib/python3.7/dist-packages/numpy/lib/npyio.py", line 962, in loadtxt
    fh = np.lib._datasource.open(fname, 'rt', encoding=encoding)
  File "/usr/local/lib/python3.7/dist-packages/numpy/lib/_datasource.py", line 266, in open
    return ds.open(path, mode, encoding=encoding, newline=newline)
  File "/usr/local/lib/python3.7/dist-packages/numpy/lib/_datasource.py", line 624, in open
    raise IOError("%s not found." % path)
OSError: Datasets/data1.txt not found.
https://ideone.com/Rpjj0v
language:
Python 3 (python  3.12)
created:
5 years ago
visibility:
public

Share or Embed source code

Discover > Sphere Engine API

The brand new service which powers Ideone!

Discover > IDE Widget

Widget for compiling and running the source code in a web browser!