Ajout de deux trois trucs

This commit is contained in:
Absobel 2021-06-06 17:59:19 +02:00
parent cc69233487
commit 1276934c54
2 changed files with 110 additions and 4 deletions

8
RUN.py
View File

@ -3,13 +3,13 @@
import mnist_loader
training_data, validation_data, test_data = mnist_loader.load_data_wrapper()
print(list(training_data)[0][1])
#print(list(training_data)[0][1])
import network
#net = network.Network([784, 30, 10]) #Testé : 94,56%
#net.SGD(training_data, 10, 10, 3.0, test_data=None)
#print("Results : {} / 10000".format(net.evaluate(test_data)))
net = network.Network([784, 30, 10]) #Testé : 94,56% / 94,87%
net.SGD(training_data, 30, 10, 3.0, test_data=test_data)
print("Results : {} / 10000".format(net.evaluate(test_data)))
# net = network.Network([784, 100, 10]) #Marche mieux apparemment
# net.SGD(training_data, 30, 10, 3.0, test_data=test_data)

View File

@ -0,0 +1,106 @@
import random
import numpy as np
class Network(object):
def __init__(self, sizes):
"""sizes : [nb neurones input, nb de neurones couche 1, ..., nb de neurones couche n, nb de neurones output]
biases : seuils générés aléatoirement
weights : poids générés aléatoirement"""
self.num_layers = len(sizes)
self.sizes = sizes
self.biases = [np.random.randn(y, 1) for y in sizes[1:]]
self.weights = [np.random.randn(y, x)
for x, y in zip(sizes[:-1], sizes[1:])]
def feedforward(self, a):
for b, w in zip(self.biases, self.weights):
a = sigmoid(np.dot(w, a)+b)
return a
def SGD(self, training_data, epochs, mini_batch_size, eta,
test_data=None):
"""epochs : iterations
eta : taux d'apprentissage
test_data : s'il y en a pas le programme ne s'arrêtera pas à chaque iterations pour se tester"""
training_data = list(training_data)
n = len(training_data)
if test_data:
test_data = list(test_data)
n_test = len(test_data)
for j in range(epochs):
random.shuffle(training_data)
mini_batches = [
training_data[k:k+mini_batch_size]
for k in range(0, n, mini_batch_size)]
for mini_batch in mini_batches:
self.update_mini_batch(mini_batch, eta)
if test_data:
print("Epoch {} : {} / {}".format(j,self.evaluate(test_data),n_test))
else:
print("Epoch {} complete".format(j))
def update_mini_batch(self, mini_batch, eta):
"""Met à jour les poids et seuils grâce aux gradient descendant"""
nabla_b = [np.zeros(b.shape) for b in self.biases]
nabla_w = [np.zeros(w.shape) for w in self.weights]
for x, y in mini_batch:
delta_nabla_b, delta_nabla_w = self.backprop(x, y)
nabla_b = [nb+dnb for nb, dnb in zip(nabla_b, delta_nabla_b)]
nabla_w = [nw+dnw for nw, dnw in zip(nabla_w, delta_nabla_w)]
self.weights = [w-(eta/len(mini_batch))*nw
for w, nw in zip(self.weights, nabla_w)]
self.biases = [b-(eta/len(mini_batch))*nb
for b, nb in zip(self.biases, nabla_b)]
def backprop(self, x, y):
"""Calcul du gradient descendant"""
nabla_b = [np.zeros(b.shape) for b in self.biases]
nabla_w = [np.zeros(w.shape) for w in self.weights]
# feedforward
activation = x
activations = [x] # list to store all the activations, layer by layer
zs = [] # list to store all the z vectors, layer by layer
for b, w in zip(self.biases, self.weights):
z = np.dot(w, activation)+b
zs.append(z)
activation = sigmoid(z)
activations.append(activation)
# backward pass
delta = self.cost_derivative(activations[-1], y) * \
sigmoid_prime(zs[-1])
nabla_b[-1] = delta
nabla_w[-1] = np.dot(delta, activations[-2].transpose())
# Note that the variable l in the loop below is used a little
# differently to the notation in Chapter 2 of the book. Here,
# l = 1 means the last layer of neurons, l = 2 is the
# second-last layer, and so on. It's a renumbering of the
# scheme in the book, used here to take advantage of the fact
# that Python can use negative indices in lists.
for l in range(2, self.num_layers):
z = zs[-l]
sp = sigmoid_prime(z)
delta = np.dot(self.weights[-l+1].transpose(), delta) * sp
nabla_b[-l] = delta
nabla_w[-l] = np.dot(delta, activations[-l-1].transpose())
return (nabla_b, nabla_w)
def evaluate(self, test_data):
"""Teste le programme avec le dataset fourni"""
test_results = [(np.argmax(self.feedforward(x)), y)
for (x, y) in test_data]
return sum(int(x == y) for (x, y) in test_results)
def cost_derivative(self, output_activations, y):
"""Return the vector of partial derivatives \partial C_x /
\partial a for the output activations."""
return (output_activations-y)
def sigmoid(z):
return 1.0/(1.0+np.exp(-z))
def sigmoid_prime(z):
return sigmoid(z)*(1-sigmoid(z))