{

"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"provenance": [],
"include_colab_link": true
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
},
"language_info": {
"name": "python"
}
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "view-in-github",
"colab_type": "text"
},
"source": [
"<a href=\"https://colab.research.google.com/github/PadmaJyothi-U/Deep-
Learning/blob/main/Simple_Neural_Network.ipynb\" target=\"_parent\"><img
src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In
Colab\"/></a>"
]
},
{
"cell_type": "markdown",
"source": [
"# **Experiment-3**\n",
"\n",
"---\n",
"\n"
],
"metadata": {
"id": "lwotv4L0gMyn"
}
},
{
"cell_type": "markdown",
"source": [
"# **Simple Neural Network**"
],
"metadata": {
"id": "cqrNe3XuhZsJ"
}
},
{
"cell_type": "markdown",
"source": [
"\n",
"The simplest type of neural network is a single-layer perceptron. It
consists of one input layer and one output layer, with no hidden layers. The
perceptron takes a set of input features and produces a binary output based on a
weighted sum of the inputs. The weights are adjusted during training to optimize
the output."
],
"metadata": {
"id": "PXF2MXSXo3PD"
}
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"id": "pzyjCDomJteF"
},
"outputs": [],
"source": [
"import numpy as np"
]
},
{
"cell_type": "code",
"source": [
"class NeuralNetwork():\n",
" \n",
" def __init__(self):\n",
" # seeding for random number generation\n",
" np.random.seed(1)\n",
" \n",
" #converting weights to a 3 by 1 matrix with values from -1 to 1
and mean of 0\n",
" self.synaptic_weights = 2 * np.random.random((3, 1)) - 1\n",
"\n",
" def sigmoid(self, x):\n",
" #applying the sigmoid function\n",
" return 1 / (1 + np.exp(-x))\n",
"\n",
" def sigmoid_derivative(self, x):\n",
" #computing derivative to the Sigmoid function\n",
" return x * (1 - x)\n",
"\n",
" def train(self, training_inputs, training_outputs,
training_iterations):\n",
" \n",
" #training the model to make accurate predictions while adjusting
weights continually\n",
" for iteration in range(training_iterations):\n",
" #siphon the training data via the neuron\n",
" output = self.think(training_inputs)\n",
"\n",
" #computing error rate for back-propagation\n",
" error = training_outputs - output\n",
" \n",
" #performing weight adjustments\n",
" adjustments = np.dot(training_inputs.T, error *
self.sigmoid_derivative(output))\n",
"\n",
" self.synaptic_weights += adjustments\n",
"\n",
" def think(self, inputs):\n",
" #passing the inputs via the neuron to get output \n",
" #converting values to floats\n",
 " \n",
" inputs = inputs.astype(float)\n",
" output = self.sigmoid(np.dot(inputs, self.synaptic_weights))\n",
" return output\n"
],
"metadata": {
"id": "mmFVZrNxJvCS"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"if __name__ == \"__main__\":\n",
" #Intialise a single neuron neural network.\n",
" neural_network = NeuralNetwork()\n",
" print(\"Random starting synaptic weights: \")\n",
" print(neural_network.synaptic_weights)\n",
"\n",
" # The training set. We have 4 examples, each consisting of 3 input
values\n",
" # and 1 output value.\n",
" training_set_inputs = array([[0, 0, 1], [1, 1, 1], [1, 0, 1], [0, 1,
1]])\n",
" training_set_outputs = array([[0, 1, 1, 0]]).T\n",
"\n",
" # Train the neural network using a training set.\n",
" # Do it 10,000 times and make small adjustments each time.\n",
" neural_network.train(training_set_inputs, training_set_outputs, 10000)\
n",
"\n",
" print(\"New synaptic weights after training: \")\n",
" print(neural_network.synaptic_weights)\n",
"\n",
" # Test the neural network with a new situation.\n",
" print(\"Considering new situation [1, 0, 0] -> ?: \")\n",
" print(neural_network.think(array([1, 0, 0])))"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "sMasE_W3gElA",
"outputId": "a6b2f8d5-5262-408b-e057-1631c6ffbaaa"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Random starting synaptic weights: \n",
"[[-0.16595599]\n",
" [ 0.44064899]\n",
" [-0.99977125]]\n",
"New synaptic weights after training: \n",
"[[ 9.67299303]\n",
" [-0.2078435 ]\n",
 " [-4.62963669]]\n",
"Considering new situation [1, 0, 0] -> ?: \n",
"[0.99993704]\n"
]
}
]
},
{
"cell_type": "code",
"source": [],
"metadata": {
"id": "d70_WGP7gInI"
},
"execution_count": null,
"outputs": []
}
]
}