Implement the forward propagation algorithm for a simple neural network with one hidden layer. The neural network has the following specifications: $1.$ Input layer with $3$ features. $2.$ Hidden layer with $4$ units, using the ReLU activation function. $3.$ Output layer with $2$ units, using the softmax activation function. Write a Python function forward$\_$propagation that takes an input array X of shape $($m$,3),$ where m is the number of examples, and returns the output predictions of shape $($m$,2).$ Assume the weights and biases of the neural network are pre$-$defined. Notes $($more detailed explanation of the question$)$: The forward propagation algorithm for a simple neural network with one hidden layer involves passing the input data through the network to make predictions. This neural network has three layers: an input layer with $3$ features, a hidden layer with $4$ units $($neurons$),$ and an output layer with $2$ units. Each unit in the hidden layer uses the ReLU activation function, while each unit in the output layer uses the softmax activation function. The purpose of the forward propagation algorithm is to take the input data and compute the predicted output of the neural network. The input data comes in the form of an array called X$,$ where each row represents an example, and there are three columns representing the three features. The goal is to calculate the predictions for each example and return the results in an array of shape $($m$,2),$ where m is the number of examples, and $2$ represents the two units in the output layer $($the classes for the binary classification problem$).$ To implement this algorithm in Python, you can create a function called forward$\_$propagation. This function takes the input array X as its input, and you should assume that the weights and biases of the neural network have been pre$-$defined. These weights and biases determine how the input data will be transformed as it passes through the neural network to produce the predictions. The steps involved in forward propagation are as follows:

Take the input X and compute the values of the hidden layer units using the ReLU activation function.

Use the weights and biases of the connections between the input layer and the hidden layer to calculate the values of the hidden units.

Once the values of the hidden units are calculated, apply the softmax activation function to calculate the predictions of the output layer.

Return the predictions for each example in the form of an array of shape $($m$,2).$

After implementing the forward$\_$propagation function, you can use it to make predictions on new data using the pre$-$defined weights and biases of the neural network.

Step $1.$ Import numpy as np$.*$Nothing to change in the code below.$*$ Step $2.$ Define the weights and biases in your forward propagation. We have predefined weights and biases below. $*$Nothing to change in the code below. Step $3.$ Enter different numbers $($ranging from $1-10)$ in the arrays $(3$ each$)$ below to see what the prediction will be$.$ Feel free to experiment by repeating an array to see what happens with the prediction.