import numpy as np
from sklearn.neural_network import  MLPClassifier

class AlgorithmMLP():
    # method: AlgorithmKNN::constructor
    #
    # arguments:
    #  win_input: GUI input display
    #  win_output: GUI output display
    #  win_log: GUI process log
    #
    # return: none
    #
    def __init__(self, win_input, win_output, win_log, nlayers, solver):
        # create class data
        #
        AlgorithmMLP.__CLASS_NAME__ = self.__class__.__name__

        # copy the inputs into class data
        #
        self.input_d = win_input
        self.output_d = win_output
        self.log_d = win_log
        self.nlayers = nlayers
        self.solver = solver

    # method: AlgorithmKNN::initialize
    #
    # arguments: None
    #
    # return: True
    #
    # initialize variables for KNN
    #
    def initialize(self, data):
        # initialize variables
        #
        self.data = data
        self.classes = len(self.data)
        self.mlp = MLPClassifier(hidden_layer_sizes=self.nlayers,
                                 solver=self.solver)
        self.X = np.empty((0, 0))
        return True

    # method: AlgorithmKNN::run_algo
    #
    # arguments: None
    #
    # return: True
    #
    # run algorithm steps
    #
    def run_algo(self, data):
        # calc everything and display stats
        self.initialize(data)
        self.train()
        return True

    def create_labels(self):
        labels = []
        count = 0
        d = self.input_d.class_info
        for i in d:
            total_samples = len(d[i][1])
            labels = labels + [count] * total_samples
            count += 1

        labels = np.array(labels)
        return labels

    def train(self):
        labels = self.create_labels()
        data = np.vstack((self.data))
        self.mlp.fit(data,labels)


    def predict(self, ax, X):
        X = np.concatenate(X, axis=0)
        X = np.reshape(X, (-1, 2))
        res = (ax.canvas.axes.get_xlim()[1] -
               ax.canvas.axes.get_ylim()[0]) / 100

        x_min, x_max = X[:, 0].min() - 1, X[:, 0].max() + 1
        y_min, y_max = X[:, 1].min() - 1, X[:, 1].max() + 1
        xx, yy = np.meshgrid(np.arange(x_min, x_max, res),
                             np.arange(y_min, y_max, res))
        Z = self.mlp.predict(np.c_[xx.ravel(), yy.ravel()])
        Z = Z.reshape(xx.shape)

        return xx, yy, Z

    def prediction_classifier(self, data):
        data = np.array([data])
        prediction = self.mlp.predict(data)
        return prediction[0]