Last active
August 29, 2015 14:14
-
-
Save tejaskhot/3eb686adea94705fb885 to your computer and use it in GitHub Desktop.
Perceptron Learning Algorithm
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import numpy as np | |
import random | |
unit_step = lambda x: 0 if x < 0 else 1 | |
## create dummy data | |
training_data = [ (np.array([0,0,1]), 0), (np.array([0,1,1]), 1), (np.array([1,0,1]), 1), (np.array([1,1,1]), 1), ] | |
w = np.random.rand(3) | |
errors = [] | |
learning_rate = 0.2 | |
n = 100 | |
for i in xrange(n): | |
x, expected = random.choice(training_data) | |
result = np.dot(w, x) | |
error = expected - unit_step(result) | |
errors.append(error) | |
w += learning_rate * error * x | |
for x, _ in training_data: | |
result = np.dot(x, w) | |
print("{}: {} -> {}".format(x[:2], result, unit_step(result))) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import numpy as np | |
import random | |
class Perceptron: | |
def __init__(self): | |
self.errors = [] | |
self.w = np.random.rand(3) | |
def train(self, X, alpha, num_iters): | |
self.training_data = X | |
for i in xrange(num_iters): | |
x, expected = random.choice(self.training_data) | |
result = np.dot(self.w, x) | |
error = expected - self.unit_step(result) | |
self.errors.append(error) | |
self.w += alpha * error * x | |
def unit_step(self, val): | |
return 0 if val < 0 else 1 | |
def test(self): | |
for x, _ in self.training_data: | |
val = np.dot(x, self.w) | |
print("{}: {} -> {}".format(x[:2], val, self.unit_step(val))) | |
training_data = [ (np.array([0,0,1]), 0), (np.array([0,1,1]), 1), (np.array([1,0,1]), 1), (np.array([1,1,1]), 1), ] | |
p = Perceptron() | |
p.train(training_data, 0.2, 100) | |
p.test() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment