import numpy as np
import matplotlib.pyplot as plt
import matplotlib.colors as mclr

from keras import layers
from keras import models

def genData(size=500):
    size1 = size//2
    size2 = size - size1

    t1 = np.random.rand(size1)
    x1 = np.asarray([i * np.math.cos(i*2*np.math.pi) + (np.random.rand(1)-1)/2*i for i in t1])
    y1 = np.asarray([i * np.math.sin(i*2*np.math.pi) + (np.random.rand(1)-1)/2*i for i in t1])
    data1 = np.hstack((x1, y1))
    label1 = np.zeros([size1, 1])
    div1 = round(size1*0.8)

    t2 = np.random.rand(size2)
    x2 = np.asarray([-i * np.math.cos(i*2*np.math.pi) + (np.random.rand(1)-1)/2*i for i in t2])
    y2 = np.asarray([-i * np.math.sin(i*2*np.math.pi) + (np.random.rand(1)-1)/2*i for i in t2])
    data2 = np.hstack((x2, y2))
    label2 = np.ones([size2, 1])
    div2 = round(size2*0.8)

    div = div1 + div2
    order = np.random.permutation(div)

    train_data = np.vstack((data1[:div1], data2[:div2]))
    test_data = np.vstack((data1[div1:], data2[div2:]))
    train_label = np.vstack((label1[:div1], label2[:div2]))
    test_label = np.vstack((label1[div1:], label2[div2:]))
    return (train_data[order, :], train_label[order, :]), (test_data, test_label)


def drawResults(data, label, prediction):
    p_label = np.array([round(x[0]) for x in prediction])
    plt.scatter(data[:, 0], data[:, 1], s=30, c=label[:, 0], cmap=mclr.ListedColormap(['red', 'blue']))
    plt.scatter(data[:, 0], data[:, 1], s=10, c=p_label, cmap=mclr.ListedColormap(['red', 'blue']))
    plt.grid()
    plt.show()


(train_data, train_label), (test_data, test_label) = genData()

#В данном месте необходимо создать модель и обучить ее
model = models.Sequential()
model.add(layers.Dense(16, activation='relu', input_shape=(2,)))
model.add(layers.Dense(16, activation='relu'))
model.add(layers.Dense(16, activation='relu'))
model.add(layers.Dense(1, activation='sigmoid'))

model.compile(optimizer='rmsprop', loss='binary_crossentropy', metrics=['accuracy'])
H = model.fit(train_data, train_label, epochs=40, batch_size=10, validation_data=(test_data, test_label))

#Получение ошибки и точности в процессе обучения
loss = H.history['loss']
val_loss = H.history['val_loss']
acc = H.history['acc']
val_acc = H.history['val_acc']
epochs = range(1, len(loss) + 1)

#Построение графика ошибки
plt.plot(epochs, loss, 'bo', label='Training loss')
plt.plot(epochs, val_loss, 'b', label='Validation loss')
plt.title('Training and validation loss')
plt.xlabel('Epochs')
plt.ylabel('Loss')
plt.legend()
plt.show()

#Построение графика точности
plt.clf()
plt.plot(epochs, acc, 'bo', label='Training acc')
plt.plot(epochs, val_acc, 'b', label='Validation acc')
plt.title('Training and validation accuracy')
plt.xlabel('Epochs')
plt.ylabel('Accuracy')
plt.legend()
plt.show()

#Получение и вывод результатов на тестовом наборе
results = model.evaluate(test_data, test_label)
print(results)

#Вывод результатов бинарной классификации
all_data = np.vstack((train_data, test_data))
all_label = np.vstack((train_label, test_label))
pred = model.predict(all_data)
drawResults(all_data, all_label, pred)