#Neural Network in Python (tested with version 3.4.0) ##Updates Added support for multiple activation functions such as:

• Linear (f(x) = x)
• Sigmoid (f(x) = 1 / (1 + e ^ -x))
• Step (f(x) = 1 {if x > 0} otherwise f(x) = 0)
• Tanh (f(x) = tanh(x))

Fixed the derivation of the tanh activation function. Fixed broken logic behind the step function. Fixed File I/O

##Usage

from layer import Layer
from neuron import Neuron
from network import Network

network = Network()
network.add_layer(10, 20, Network.ACTIVATION_STEP) # Hidden Layer, 10 Neurons, 20 inputs
network.add_layer(2,  10, Network.ACTIVATION_STEP) # Output Layer,  2 Neurons, 10 inputs

# Simulate black and white images
# 0 - Black
# 1 - White
zero = [
    0, 1, 1, 0,
    1, 0, 0, 1,
    1, 0, 0, 1,
    1, 0, 0, 1,
    0, 1, 1, 0
]

one = [
    0, 0, 1, 0,
    0, 0, 1, 0,
    0, 0, 1, 0,
    0, 0, 1, 0,
    0, 0, 1, 0
]

two = [
    0, 1, 1, 0,
    1, 0, 0, 1,
    0, 0, 1, 0,
    0, 1, 0, 0,
    1, 1, 1, 1
]

three = [
    1, 1, 1, 1,
    0, 0, 0, 1,
    0, 1, 1, 1,
    0, 0, 0, 1,
    1, 1, 1, 1
]

# Set some quick properties for the upcoming training session
ITERATIONS = 500  # Number of iterations per training session
LEARN_RATE = 0.1  # The rate the network learns on each iteration
THRESHOLD  = 0.01 # If this precision is reached, the training session is instantly complete

# Perform a quick training session
for i in range(0, ITERATIONS, 1):
    error = 0
    
    #        Examples      Inputs   Outputs Learn Rate
    error += network.train(zero,    [0, 0], LEARN_RATE)
    error += network.train(one,     [0, 1], LEARN_RATE)
    error += network.train(two,     [1, 0], LEARN_RATE)
    error += network.train(three,   [1, 1], LEARN_RATE)

    # Check the error
    if error < THRESHOLD:
        break

# Process the outputs
outputs = network.process([
    0, 1, 1, 0,
    1, 0, 0, 1,
    0, 0, 1, 0,
    0, 1, 0, 0,
    1, 1, 1, 1
])

# Print the decimal equivalent of the network's output (Expected: 2)
print(sum([(2 ** i) * outputs[-i - 1] for i in range(0, len(outputs), 1)]))

##Credits Made by Jayrese Heslop