import numpy as np

# two inputs [sleep,study]

X = np.array(([2, 9], [1, 5], [3, 6]), dtype=float) 


 # one output [Expected % in Exams]  # Labels(Marks obtained)

y = np.array(([92], [86], [89]), dtype=float)


# maximum of X array longitudinally

X = X/np.amax(X,axis=0) # Normalize # maximum of X array


y = y/100  # max test score is 100


#Sigmoid Function

def sigmoid (x):

    return 1/(1 + np.exp(-x))


#Derivative of Sigmoid Function

def derivatives_sigmoid(x):

    return x * (1 - x)


#Variable initialization

epoch=2000 #Setting training iterations

lr=0.1 #Setting learning rate

inputlayer_neurons = 2 #number of features in data set

hiddenlayer_neurons = 3 #number of hidden layers neurons

output_neurons = 1 #number of neurons at output layer


#weight and bias-Random initialization

wh=np.random.uniform(size=(inputlayer_neurons,hiddenlayer_neurons)) 

#weight of the link from input node to hidden node  2*3


# bias of the link from input node to hidden node  1*3

bh=np.random.uniform(size=(1,hiddenlayer_neurons)) 


#weight of the link from hidden node to output node 1*1

wout=np.random.uniform(size=(hiddenlayer_neurons,output_neurons)) 


#bias of the link from hidden node to output node

bout=np.random.uniform(size=(1,output_neurons)) 



#draws a random range of numbers uniformly of dim x*y

for i in range(epoch):


#Forward Propogation

    hinp1=np.dot(X,wh)  # Dot product + bias 

    hinp=hinp1 + bh

    

    hlayer_act = sigmoid(hinp)  # Activation function 

    

    outinp1=np.dot(hlayer_act,wout)

    outinp= outinp1+ bout

    output = sigmoid(outinp)


#Backpropagation

# Error at Output layer

    EO = y-output  # Errj=Oj(1-Oj)(Tj-Oj)

    outgrad = derivatives_sigmoid(output)

    d_output = EO* outgrad

    EH = d_output.dot(wout.T)    # .T means transpose


#how much hidden layer weights contributed to error

    hiddengrad = derivatives_sigmoid(hlayer_act)

    d_hiddenlayer = EH * hiddengrad


# dotproduct of nextlayererror and currentlayerop

wout += hlayer_act.T.dot(d_output) *lr

wh += X.T.dot(d_hiddenlayer) *lr


print("Input: \n" + str(X)) 

print("Actual Output: \n" + str(y))

print("Predicted Output: \n" ,output)