__author__ = "greg" import loader import neural_network training_data, validation_data, test_data = loader.load_data_wrapper() net = neural_network.Network([784, 30, 10]) net.SGD(training_data, 30, 10, 3.0, test_data=test_data)
# ---------------------- # - read the input data: import loader import matrix as mtx training_data, validation_data, test_data = loader.load_data_wrapper() training_data = list(training_data) net = mtx.Matrix([784, 100, 10]) net.learn(training_data, 30, 10, 5.0, test_data)
def train_network(net): # Load training data training_data, validation_data, test_data = loader.load_data_wrapper() # Train neural network net.SGD(training_data, 30, 10, 3.0, test_data=test_data)
def evaluate(self, data): test_results = [(np.argmax(self.feedforward(x)), np.argmax(y)) for (x, y) in data] return sum(int(x == y) for (x, y) in test_results) def save(self, filename): data = { "sizes": self.sizes, "weights": [w.tolist() for w in self.weights], "biases": [b.tolist() for b in self.biases], "cost": str(self.cost.__name__) } f = open(filename, "w") json.dump(data, f) f.close() def sigmoid(z): return 1.0 / (1.0 + np.exp(-z)) def sigmoid_prime(z): return sigmoid(z) * (1 - sigmoid(z)) if __name__ == '__main__': tr, va, te = loader.load_data_wrapper() network = Network([784, 30, 10]) network.SGD(tr, 30, 10, 0.5, 5.0, va, True, True, True, True) network.save("improved_network_results.json")
import numpy as np from keras.models import Sequential from keras.layers import Dense from loader import load_data_wrapper train, valid, test = load_data_wrapper() model = Sequential() model.add(Dense(784, input_dim=784, activation='relu')) model.add(Dense(30, activation='relu')) model.add(Dense(10, activation='softmax')) model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy']) print("finished compiling") inp = [] out = [] count = 0 for i, o in train: if len(inp) > 8000: break base_i = [] base_o = [] for item in i: base_i.append(item[0]) for item in o: