Keras_XOR_NN.py

from keras.models import Sequential
from keras.layers.core import Dense, Dropout, Activation
from keras import losses
import numpy as np

X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([[0], [1], [1], [0]])

model = Sequential()

model.add(Dense(2, input_dim=2, activation='relu'))
model.add(Dense(1, activation='sigmoid'))

model.compile(loss='mean_squared_error', optimizer='adam', metrics=['accuracy'])

model.fit(X, y, epochs=500, batch_size=10)
_, accuracy = model.evaluate(X, y)
print('Accuracy: %.2f' % (accuracy * 100))

for layer in model.layers:
    weights = layer.get_weights()  # list of numpy arrays
    print(weights)

x_new = np.array([[0, 1]])
predict = model.predict_classes(x_new)
print(predict)
#Anther way 

X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
y = np.array([[0], [1], [1], [0]])

print(X.shape)

x = Input(shape=(2,))
h = Dense(3, activation='relu')(x)
r = Dense(1, activation='sigmoid')(h)

model = Model(inputs=x, outputs=r)

model.compile(optimizer='adam', loss='mse')

history = model.fit(X, y, batch_size=4, epochs=5, verbose=1, validation_data=(X, y))

plt.plot(list(history.history.values())[0], 'k-o')
plt.show()