from neural import dataset as db
from neural import graphic as gp
from sklearn.model_selection import KFold
from sklearn.model_selection import train_test_split as split
import numpy as np
REALIZATIONS = 20
N_SPLITS = 5
database = db.dataset()
x_data, y_data = database.regressionDatabase()
graph = gp.Graph()
# j = nn.test(x, activation="linear")
grid = sg.SearchGrid(type="regression")
RMSE, MSE, WEIGTHS = [], [], []
for i in range(REALIZATIONS):
x_train, x_test, y_train, y_test = split(x_data, y_data, test_size=0.2)
nn = network.mlp(inputs=x_train,
outputs=y_train,
outLayer=1,
hiddenLayer=4,
eta=0.01)
nn.trainer(1000, activation="linear")
print(" > -----------------------------------------------")
print(" > Realizações: ", i + 1)
from neural import cross_validation as cv
from neural import dataset as db
from neural import search_grid as sg
import numpy as np
from sklearn.model_selection import KFold
from sklearn.model_selection import train_test_split as split
N_SPLITS = 5
REALIZATIONS = 20
# import dataset
database = db.dataset()
x_data, y_data = database.dermatology()
# search grid
grid = sg.SearchGrid()
print("## Problema: DERMATOLOGY ")
print("## Treinamento: KFold - K:", N_SPLITS)
for epoch in grid.getEpochs():
for eta in grid.getEtas():
for neuron in grid.getNeurons():
x_train, x_test, y_train, y_test = split(x_data,
y_data,
test_size=0.2)
TAXAS_FINAL = []
print("---------------------------------------------------")
print(" > Epoch", epoch, "Eta:", eta, "Neurons:", neuron)