for i in range(len(data[0, :])):
_max = np.max(data[:, i])
_min = np.min(data[:, i])
data[:, i] = (data[:, i] - _min) / (_max - _min)
trainPortion = int(len(data) * 66 / 100)
trainData = data[:trainPortion]
testData = data[trainPortion:]
test_dataset = testData[:, :-1].tolist()
test_target = testData[:, -1:].tolist()
dataset = trainData[:, :-1].tolist()
target = trainData[:, -1:].tolist()
myNet5 = net.Dense()
myNet5.AddLayer(13, isInput=True)
myNet5.AddLayer(5, activationFunction=af.Sigmoid)
myNet5.AddLayer(1, activationFunction=af.Sigmoid)
myNet20 = net.Dense()
myNet20.AddLayer(13, isInput=True)
myNet20.AddLayer(20, activationFunction=af.Sigmoid)
myNet20.AddLayer(1, activationFunction=af.Sigmoid)
start_time = time.time()
myNet5.Train(dataset,
target,
iterationCount=20,
learningRateStart=.6,
learningRateEnd=.0,
import Network as net
import ActivationFunction as af
myNet = net.Dense()
myNet.AddLayer(2, isInput=True)
myNet.AddLayer(3, activationFunction=af.Sigmoid)
myNet.AddLayer(1, activationFunction=af.Sigmoid)
dataset = [[5, 1], [8, 2], [9, .5], [7, 1.2], [.5, 8], [1.2, 9.5], [.7, 7],
[1.5, 6]]
target = [[0], [0], [0], [0], [1], [1], [1], [1]]
myNet.Train(dataset, target, 1000, .2)
print(myNet.GetOutput([6.5, 1.5]))
print(myNet.GetOutput([.1, 10]))