# # training data
# x_train = np.array([[[0,0]], [[0,1]], [[1,0]], [[1,1]]])
# one_hot_encoded_y_train = np.array([[[1, 0]], [[0, 1]], [[0, 1]], [[0, 1]]])
# y_train =[0, 1, 1, 1]
# this line is used to catch the errors arising from numpy.
np.seterr(all='raise')
input_number = x_train.shape[2]
output_number = 6
size_of_hidden_layer = 10
neural_network = NeuralNetwork(cross_entropy, cross_entropy_prime)
neural_network.add_layer(
FCLayer(input_number, size_of_hidden_layer, diminishing_factor=10))
neural_network.add_layer(ActivationLayer(swish, swish_prime))
neural_network.add_layer(FCLayer(size_of_hidden_layer, output_number))
neural_network.add_layer(ActivationLayer(softmax, softmax_prime))
neural_network.fit(x_train,
one_hot_encoded_y_train,
epoch_number=10,
initial_learning_rate=0.5,
decay=0.01)
out = neural_network.predict(x_train)
predictions = argmax(out)
print("confusion matrix:", confusion_matrix(y_train, predictions), sep="\n")
print("accuracy: ", accuracy_score(y_train, predictions))
print("end")
if __name__ == '__main__':
X, y = gen_mult_ser(3000)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4)
clf = NeuralNetwork(optimizer=optimizer, loss=CrossEntropy)
clf.add(RNN(10, activation="tanh", bptt_trunc=5, input_shape=(10, 61)))
clf.add(Activation('softmax'))
tmp_X = np.argmax(X_train[0], axis=1)
tmp_y = np.argmax(y_train[0], axis=1)
print("Number Series Problem:")
print("X = [" + " ".join(tmp_X.astype("str")) + "]")
print("y = [" + " ".join(tmp_y.astype("str")) + "]")
print()
train_err, _ = clf.fit(X_train, y_train, n_epochs=500, batch_size=512)
y_pred = np.argmax(clf.predict(X_test), axis=2)
y_test = np.argmax(y_test, axis=2)
accuracy = np.mean(accuracy_score(y_test, y_pred))
print(accuracy)
print()
print("Results:")
for i in range(5):
tmp_X = np.argmax(X_test[i], axis=1)
tmp_y1 = y_test[i]
tmp_y2 = y_pred[i]
print("X = [" + " ".join(tmp_X.astype("str")) + "]")
print("y_true = [" + " ".join(tmp_y1.astype("str")) + "]")
print("y_pred = [" + " ".join(tmp_y2.astype("str")) + "]")
print()