# the three different parameters has any significant impact on predicting the # value of any of the 3 parameters xDatas = [yArray, yArray, (yArray - np.mean(yArray)) / np.std(yArray) - uArray] yDatas = [taus, kps, thetas] m = model(nstep) modelList = [] uSum = np.sum(uArray, axis=0) plt.figure() plt.plot(np.linspace(0, 100, 100), uSum) for j in range(0, len(xDatas)): xData = xDatas[j] yData = yDatas[j] x_train, x_val, y_train, y_val, numDim = sig.preprocess(xData, yData) model = m.model_1() print("Fit model on training data") history = model.fit( x_train, y_train, batch_size=16, epochs=100, # We pass some validation for # monitoring validation loss and metrics # at the end of each epoch validation_data=(x_val, y_val), ) modelList.append(model)