Using the monks data set, monks$-1.$train and monks$-1.$test that has six attributes and binary class labels, please write python code for a jupyter notebook that satisfies the following.

$1.$ Implement a three layer neural network, layer one $8$ nodes snd a hyperbolic tangent activation function, layer two has $6$ nofed snd a relu activation function, layer $3$ has one node and a sigmoid activation function. Can use either the sequential and dense layer grom keras or sckit learn MLPClassifier or implement your own neural network function.

$2.$ Use the cross entropy function as the loss function for logistic regression

$3.$ Set the learning rate as $0\mathrm{.}01$ or change to other values if you find $0\mathrm{.}01$ is not suitable for this problem.

$4.$ For the two optimizers: stochastic gradient descent and Adam, record the testing accuracy of these two optimizers with respect to the epochs number

An example of $2$ layer neural network using keras with $12$ nodes in hidden layer: from keras.models import Sequential from keras.layers import Dense model $=$ Sequential$()$

model.add$($Dense$(12,$ input$\_$dim$=8,$ activation$=$'relu'$))$

model.add$($Dense$(1,$ activation$=$'sigmoid'$))$

model.compile$($loss$=$'binary$\_$crossentropy', optimizer$=\text{'}$sgd$\text{'})$

model.fit$($X$,$y$)$