import neuralpy net = neuralpy.Network(2, 10, 8, 1) neuralpy.output(net.feedforward([1, 1])) neuralpy.output(net.feedforward([0, 1])) neuralpy.output(net.feedforward([1, 0])) neuralpy.output(net.feedforward([0, 0])) datum_1 = ([1, 1], [0]) datum_2 = ([1, 0], [1]) datum_3 = ([0, 1], [1]) datum_4 = ([0, 0], [0]) training_data = [datum_1, datum_2, datum_3, datum_4] epochs = 300 learning_rate = 3 net.train(training_data, epochs, learning_rate, monitor_cost=True) neuralpy.output() neuralpy.output(net.feedforward([1, 1])) neuralpy.output(net.feedforward([0, 1])) neuralpy.output(net.feedforward([1, 0])) neuralpy.output(net.feedforward([0, 0])) net.show_costs()
# |_______|_______|___________| # | F | T | T | # |_______|_______|___________| # | T | F | T | # |_______|_______|___________| # | F | F | F | # |_______|_______|___________| # # In our network, 1 will represent True and # 0 will represent False # import neuralpy # set up a basic neural network with a 2-node input layer, # one 3-neuron hidden layer, and one 1-neuron output layer net = neuralpy.Network(2, 3, 1) # here is some arbitrary input that we will use to test x = [1, 1] out = net.feedforward(x) neuralpy.output(out) # here is our training_data the reflects the truth table # in the header of this file datum_1 = ([1, 1], [1]) datum_2 = ([1, 0], [1]) datum_3 = ([0, 1], [1]) datum_4 = ([0, 0], [0]) training_data = [datum_1, datum_2, datum_3, datum_4]
print y.head(10)
print x.describe()
print y.describe()
"""
# organize the data to be used by the DNN
count = 0
dataSet = [([x.ix[count]],[y.ix[count]])] # list containing tuples of lists
count =1
while (count<sample_size ):
#print " Working on data item : ",count
dataSet = (dataSet +[([x.ix[count]] ,[y.ix[count]])])
count = count + 1
# Get the neural network library
import neuralpy
fit = neuralpy.Network(1,3,7,1) # each argument is the number of neurons in the layer. The 1st one is the input, last one is the output
epochs = 100
learning_rate = 1
print " fitting model right now "
fit.train( dataSet , epochs , learning_rate )
# Assess model performance
count = 0
pred =[]
while ( count < sample_size ):
out = fit.forward(x[count]) # the neural network ouput
print(" Obs : ",count +1," y = ",round(y[ count ] ,4) ,
" prediction = ", round(pd.Series(out) ,4))
pred.append(out)
count = count + 1
# print(x.head(10))
# print(y.head(10))
# View information about the data: count mean std min 25% 50% 75% max
# print(x.describe())
count = 0
# dataSet is a list
dataSet = [([x.ix[count]]), ([y.ix[count]])]
count = 1
while (count < sample_size):
print("Working on data item: ", count)
dataSet = (dataSet + [([x.ix[count, 0]])], [([y.ix[count, 0]])])
count = count + 1
fit = neuralpy.Network(1, 3, 7, 1)
epochs = 100
learning_rate = 1
print("fitting model right now")
fit.train(dataSet, epochs, learning_rate)
count = 0
pred = []
while (count < sample_size):
out = fit.forward(x[count])
print("Obs: ", count + 1, "y = ", round(y[count], 4), "prediction = ",
round(pd.Series(out), 4))
pred.append(out)
count = count + 1
import neuralpy layers = [2, 10, 8, 1] net = neuralpy.Network(layers) #neuralpy will automatically generate random incoming weights and biases for each processing layer. datum_1 = ([1, 1], [0]) datum_2 = ([1, 0], [1]) datum_3 = ([0, 1], [1]) datum_4 = ([0, 0], [0]) training_data = [datum_1, datum_2, datum_3, datum_4] epochs = 300 learning_rate = 3 #monitor_cost tracks the cost of every epoch net.train(training_data, epochs, learning_rate) print net.forward([1,0]) print net.forward([0,0]) print net.forward([0,1]) print net.forward([1,1])
import pandas as pd
import neuralpy as ne
import random
random.seed(2016)
sample_size=50
sample = pd.Series(random.sample(range(-10000,10000),sample_size))
x=sample/10000
y=x**2
count = 0
dataSet = [([x.ix[count]],[y.ix[count]])]
count = 1
while (count < sample_size):
# print "Working on data item: ", count
dataSet = (dataSet+[([x.ix[count,0]], [y.ix[count]])])
count += 1
fit = ne.Network(1, 6, 12, 1)
epochs = 5000
learing_rate = 1
print "Fitting model right now"
fit.train(dataSet, epochs, learing_rate)
count = 0
pred = []
while(count < sample_size):
out = fit.forward(x[count])
print 'Obs:' + str((count+1)) + '\t' + 'y=' + str(round(y[count],4)) + '\t' + 'prediction=' + str(round(pd.Series(out),4))
pred.append(out)
count += 1