number of neurons =	{16, 32, 64, 128, 256, 512, 1024}
number of layers =	{2, 3, 4, 5, 6}
activation function =	{relu, softplus, linear, sigmoid, tanh, softsign, softmax}
optimizer =	{rmsprop, adam, sgd, adagrad, adadelta, adama, nadam}
lr =	{0.1, 0.01, 0.001, 0.0001}
loss =	{binary cross-entropy, mean squared error}
dropout =	{0, 0.2, 0.3, 0.5}