Scribd
Upload
10 views4 pages

Tests

Tests for ANN

Uploaded by

oscar.audzipetras
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
10 views4 pages

Tests

Tests for ANN

Uploaded by

oscar.audzipetras
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 4

from layer import Layer

from network import Network


from termcolor import colored
from typing import List

class Tests:
def __init__(self):

"""
Initializes modules required for the tests
"""

self.layer = Layer(2, 3)
self.network = Network([2, 3, 2])

def test_layer_output_calculation(self):

"""
Test for the calculate outputs function for the individual layers.
Currently configured for Sigmoid Squishification activation function
"""

# Setting layer weights and biases to predetermined values

self.layer.weights = [
[1.0, 0.5],
[0.75, 1.5],
[0.0, 2.0]
]

self.layer.biases = [0.5, 1.0, 1.5]

# Setting the inputs of the layer

inputs = [2.0, 0.5]

# Calculating what the module thinks the outputs are

outputs = self.layer.calculate_outputs(inputs)

# Rounding outputs to 5 decimal places

for i in range(len(outputs)):
outputs[i] = round(outputs[i], 5)

# Defining precomputed, correct outputs

expected_outputs = [0.93991, 0.96267, 0.92414]

# Checking if the output of the function in question and precomputed,


correct outputs match

self.print_test_results("LAYER CALCULATE OUTPUTS", outputs,


expected_outputs)

def test_network_feed_inputs(self):

"""
Test for feeding the inputs to the network.
"""

# Defining inputs to be fed into the network

inputs = [2.0, 1.0]

# Feeding inputs into the network

self.network.feed_inputs(inputs)

# Checking if the inputs have been fed to the network successfully

self.print_test_results("NETWORK FEED INPUTS", self.network.inputs, inputs)

def test_network_run_calculations(self):

"""
Test for the entire neural network. Currently configured for sigmoid
squishification activation function
"""

# Setting Network weights and biases to predetermined values

self.network.layers[0].weights = [
[1.0, 0.5],
[0.75, 1.5],
[0.0, 2.0]
]

self.network.layers[0].biases = [0.5, 1.0, 1.5]

self.network.layers[1].weights = [
[0.5, 1.0, 2.5],
[0.0, 0.5, -1.5],
]

self.network.layers[1].biases = [0.0, -2.0]

# Defining inputs to be fed to the network

inputs = [2.0, 0.5]

# Defining precomputed, correct values for what the function being tested
should produce

expected_outputs = [0.97686, 0.05191]

# Feeding the inputs into the network

self.network.feed_inputs(inputs)

# Storing the function being tested's values

outputs = self.network.run_calculations()

# Rounding values to 5 decimal places

for i in range(len(outputs)):
outputs[i] = round(outputs[i], 5)
# Checking if the output of the function in question, and precomputed,
correct outputs match

self.print_test_results("NETWORK RUN CALCULATIONS", outputs,


expected_outputs)

def test_activation_function(self):

"""
Test for the activation function. Currently configured to test sigmoid
squishification
"""

# Defining the various inputs that the activation function will be tested
for

x = [10.0, 0.0, 5.0, 3.2]

# Calculating the outputs of the activation function and rounding them to 5


decimal places

for i in range(len(x)):
x[i] = round(self.layer.activation_function(x[i]), 5)

# Defining precomputed and correct values for function

expectation = [0.99995, 0.5, 0.99331, 0.96083]

# Checking if the output of the function in question and precomputed,


correct outputs match

self.print_test_results("LAYER ACTIVATION FUNCTION", x, expectation)

def print_test_results(self, name: str, result, expected_result):

"""
Generic function for printing the test results of a test. Parameters in
order:
1. name : Specifies the name of the test, will be printed at the
top
2. result : The result that the functionality being tested gave
3. expected_result : The result that the functionality being tested should
give for the test to pass

If the result equals the expected result this function will print that it
passed, otherwise,
this function will print that the test failed failed.
"""

print("---------------------------------------")
if (result == expected_result):
print(colored(f"TEST : {name} : PASSED", "cyan", attrs=['bold']))
else:
print(colored(f"TEST : {name} : FAILED", "red", attrs=['bold']))

print("---------------------------------------")
print(F"EXPECTED OUTPUT : {expected_result}")
print(f"GIVEN OUTPUT : {result}")
print("---------------------------------------\n")

You might also like