Neural Network Library - Revision history

Elverion at 19:15, 17 February 2018

2018-02-17T19:15:54Z

2018-02-17T19:15:21Z

‎Examples

2018-02-17T18:56:21Z

‎NeuralNet:load()

2018-02-17T18:55:05Z

‎NeuralNet:save()

2018-02-17T18:54:13Z

‎NeuralNet:getExportTable()

2018-02-17T18:54:00Z

‎NeuralNet:getExportTable()

2018-02-17T18:50:58Z

‎NeuralNet:getResults()

2018-02-17T18:49:11Z

‎NeuralNet:backPropagation()

2018-02-17T18:40:36Z

‎NeuralNet:feed()

2018-02-17T18:31:03Z

‎Creating an instance of a neural network

← Older revision		Revision as of 19:15, 17 February 2018
Line 88:		Line 88:

	= Examples =		= Examples =
		+	==Calculating xor==
		+
		+	In this example, we will create a neural network and train it to calculate for xor. Xor (eXclusive Or) simply means that input 1 OR input 2 must be true(1) but not both at the same time.
		+	1 xor 0 is true(1)
		+	0 xor 1 is true(1)
		+	1 xor 1 is false(0)
		+	0 xor 0 is false(0)
		+
		+
		+	Since Lua already provides a way to do this, a neural network is way overkill and computationally expensive in comparison, however it should make a good example.
		+
		+	<source lang="lua">
		+	require('neuralnet/net');
		+
		+	function macro.init()
		+	net = NeuralNet(2, 2, 1);
		+
		+	for i = 1,1000 do
		+	local a = math.random(0, 1);
		+	local b = math.random(0, 1);
		+
		+	net:feed(a, b);
		+
		+	local result = a ~ b; -- ~ is the bitwise 'xor' operation
		+	net:backPropagation(result); -- Train on this data for the given input
		+	end
		+
		+	-- Our network should be trained well enough, so lets see if it works
		+	local a = math.random(0, 1);
		+	local b = math.random(0, 1);
		+	local expected = a ~ b;
		+
		+	net:feed(a, b);
		+	local results = net:getResults();
		+	local networkXor = math.floor(results[1] + 0.5); -- Since the network outputs a floating-point number, we round it
		+	printf("Our neural network thinks %d XOR %d is %d. We expect to get %d\n", a, b, networkXor, expected);
		+	end
		+	</source>

← Older revision		Revision as of 18:56, 17 February 2018
Line 83:		Line 83:

	==NeuralNet:load()==		==NeuralNet:load()==
		+	'''NeuralNet:load(filename)'''
		+
		+	Loads the neuron weights back into a network from a file that was saved with NeuralNet:save()

	= Examples =		= Examples =

@@ Line 78: / Line 78: @@
 ==NeuralNet:save()==
 ==NeuralNet:load()==
 = Examples =

@@ Line 71: / Line 71: @@
 ==NeuralNet:getExportTable()==
-'''table NeuralNet:getExportTable'''
+'''table NeuralNet:getExportTable()'''
 Returns a table containing neuron weights for the whole network. You almost certainly won't want to have to re-train your network every time you want to use it, so this function helps with saving all necessary data to restore the network to a trained state.

@@ Line 62: / Line 62: @@
 ==NeuralNet:getRecentAverageError()==
 ==NeuralNet:getResults()==
 ==NeuralNet:getExportTable()==
 ==NeuralNet:save()==

@@ Line 33: / Line 33: @@
 ==NeuralNet:feed()==
 ==NeuralNet:backPropagation()==
 ==NeuralNet:getRecentAverageError()==

@@ Line 27: / Line 27: @@
 net = NeuralNet(10, 6, 2); -- Create a neural network with 10 inputs, one hidden layer with 6 neurons, and 2 outputs.
 net2 = NeuralNet(32, 24, 16, 8); -- 32 inputs, 24 neurons in hidden layer #1, 16 neurons in hidden layer #2, and 8 outputs.
 </source>