def train(batch_size, iterate_num, learning_rate):
"""
Get the appropriate network parameters (weights, biases) by gradient method.
In the process of gradient method,
training data are choosed randomly in each step(mini-batch gradient method).
batch_size: data of this number are choosed from training data in each step
iterate_num: the number of iteration for gradient method
learning_rate: learning rate for gradient method
"""
# get training data and test data(test data are not used below.)
(x_train, t_train), (x_test, t_test) = load_mnist(normalize=True)
# initialized TwoLayerNet
network = TwoLayerNet(28 * 28, 50, 10) # each image has 28*28 pixels
# losses in each step
losses = []
for i in range(iterate_num):
# choose the training data for this step
indices = np.random.choice(len(x_train), batch_size)
x_train_batch = x_train[indices]
t_train_batch = t_train[indices]
# calculate the grad
grads = network.numerical_gradient(x_train_batch, t_train_batch)
# update the network parameters
network.params['W1'] -= learning_rate * grads['W1']
network.params['b1'] -= learning_rate * grads['b1']
network.params['W2'] -= learning_rate * grads['W2']
network.params['b2'] -= learning_rate * grads['b2']
# record loss
loss = network.loss(x_train_batch, t_train_batch)
print('loss = {0}'.format(loss))
losses.append(loss)
return network, losses
# coding: utf-8
import sys, os
from ch04.two_layer_net import TwoLayerNet
sys.path.append(os.pardir) # 为了导入父目录的文件而进行的设定
import numpy as np
import matplotlib.pyplot as plt
from dataset.mnist import load_mnist
# from two_layer_net import TwoLayerNet
# 读入数据
(x_train, t_train), (x_test, t_test) = load_mnist(normalize=True,
one_hot_label=True)
network = TwoLayerNet(input_size=784, hidden_size=50, output_size=10)
iters_num = 10000 # 适当设定循环的次数
train_size = x_train.shape[0]
batch_size = 100
learning_rate = 0.1
train_loss_list = []
train_acc_list = []
test_acc_list = []
iter_per_epoch = max(train_size / batch_size, 1)
for i in range(iters_num):
batch_mask = np.random.choice(train_size, batch_size)
x_batch = x_train[batch_mask]
import numpy as np
from dataset.mnist import load_mnist
from ch04.two_layer_net import TwoLayerNet
(x_train, t_train), (x_test, t_test) = \
load_mnist(normalize = True, one_hot_label = True)
train_loss_list = []
# Hyperparameters
iters_num = 10000
train_size = x_train.shape[0]
batch_size = 100
learning_rate = 0.1
network = TwoLayerNet(input_size=784, hidden_size=50, output_size=10)
for i in range(iters_num):
# Get mini batch
batch_mask = np.random.choice(train_size, batch_size)
x_batch = x_train[batch_mask]
t_batch = t_train[batch_mask]
# Calcuate gradient
grad = network.numerical_gradient(x_batch, t_batch)
# grad = network.gradient(x_batch, t_batch) # Improved version
# Update hyperparameters
for key in ('W1', 'b1', 'W2', 'b2'):
network.params[key] -= learning_rate * grad[key]
# # 计算梯度
# grad = network.numerical_gradient(x_batch, t_batch)
# # 按照梯度方向更新
# for key in ('W1', 'b1', 'W2', 'b2'):
# network.params[key] -= learn_num * grad[key]
# # 计算损失函数
# loss = network.loss(x_batch, t_batch)
# train_loss.append(loss)
# print(loss)
# print(train_loss)
# # 正向计算和反向计算图
# # 反向传播求导
# # 通过计算图，节点位置为： A → (x2) → 2A → (x1.1) → 2.2A 正向传播
# # 反向： 2.2 ← (x2) ← 1.1 ← (x1.1) ← 1 反向传播主要是，在结果处设置1，然后反向求，得到的2.2含义是，当A增加一个微小值ETA，则总数会增加2.2ETA，这就是导数
network = TwoLayerNet(input_size=784, hidden_size=50, output_size=10)
iters_num = 10000 # 适当设定循环的次数
train_size = x_train.shape[0]
batch_size = 100
learning_rate = 0.1
train_loss_list = []
iter_per_epoch = max(train_size / batch_size, 1)
for i in range(iters_num):
batch_mask = np.random.choice(train_size, batch_size)
x_batch = x_train[batch_mask]
t_batch = t_train[batch_mask]