## EXPLAIN MULTILAYER PERCEPTRON

Let’s understand multilayer perceptron using example-:

– As tensorflow is an open source software library by Google for numerical computation using data flow graph.

– Basic understanding of python should be good if you will use it for MLP.

### Now make a multilayer perceptron.

Multilayer perceptron is a feed forward artificial neural network model that map set of input data to set of appropriate output MLP. As MLP contains multi layer of nodes in a directed graph. In this layer are connected.

>import tensorflow as tf

> import numpy as np

>and further go with data as-:  – As it’s going to be three layer neural networks and there is one hidden layer, first layer has two modes, and the second layer is hidden layer, third layer is the output layer that has only one node as in above code. It’s going to set of weight i.e. W1 and W2 with set of bias.

– Here set of weight is going to use to connect two layers of neurons. Grid W1 is to connect input layer to hidden layer whereas the W2 is used to connect hidden layer to output layer.

– Another you have to do is create the placeholder. Placeholder is a variable to which you are going to assign value later.  – Now do the calculations with the input and set of weight and bias as-: Now calculate the cost and create optimizer function using gradient function as it improves to minimize the cost and improve output.

And then initialize the variables and start tensorflow and start creating loop and after calculate the answer -: Now test the data as-: Now run this in the terminal as-: From output you can see that cost is improving after every thousand creation and in end you can see the accuracy of 1.0, so make some changes in it as multiply by 100 to get proper answer. Now check it again. So output tells that cost is improving and accuracy is 100%, which conclude that increase in the number of filtration, then cost will decrease and accuracy will increase with better output.

Apply now for Advanced Python Training Course