# 本书链接https://tangshusen.me/Dive-into-DL-PyTorch/#/chapter03_DL-basics/3.6_softmax-regression-scratch # 3.7节 # 注释：黄文俊 # E-mail：[email protected] import torch from torch import nn from torch.nn import init import numpy as np import sys sys.path.append("..") import d2lzh_pytorch as d2l from collections import OrderedDict batch_size = 256 train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size) num_inputs = 784 num_outputs = 10 # class LinearNet(nn.Module): # def __init__(self, num_inputs, num_outputs): # super(LinearNet, self).__init__() # self.linear = nn.Linear(num_inputs, num_outputs) # def forward(self, x): # x shape: (batch, 1, 28, 28) # y = self.linear(x.view(x.shape[0], -1)) # return y # net = LinearNet(num_inputs, num_outputs) # print(net) # 使用print可以打印出网络的结构 # 本函数已保存在d2lzh_pytorch包中方便以后使用
nn.Sigmoid(),
nn.Linear(84, 10))
print('构造一个高和宽均为28的单通道数据样本，并逐层进行前向计算来查看每个层的输出形状')
X = torch.randn(size=(1, 1, 28, 28), dtype=torch.float32)
for layer in net:
X = layer(X)
print(layer.__class__.__name__, 'output shape: \t', X.shape)
print('-----------------------------------------------')
print('训练')
print('-----------------------------------------------')
print('获取数据,我们仍然使用Fashion-MNIST作为训练数据集')
# 数据
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size=batch_size,
root='../../dataset')
print(len(train_iter))
print('查看gpu设备是否可用')
# This function has been saved in the d2l package for future use
#use GPU
def try_gpu():
"""If GPU is available, return torch.device as cuda:0; else return torch.device as cpu."""
if torch.cuda.is_available():
device = torch.device('cuda:0')
else:
device = torch.device('cpu')
return device
#%% [markdown]
import torch
from torch import nn
from torch.nn import init
import numpy as np
import sys
sys.path.append("..")
import d2lzh_pytorch as d2l
#%% [markdown]
# 1.读取数据
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size,
resize=None,
root="Datasets")
#%% [markdown]
# 2.定义和初始化模型
num_inputs = 784
num_outputs = 10
class LinearNet(nn.Module):
def __init__(self, num_inputs, num_outputs):
super(LinearNet, self).__init__()
self.linear = nn.Linear(num_inputs, num_outputs)
def forward(self, x):
y = self.linear(x.view(x.shape[0], -1))
return y
@description: Softmax与分类模型
"""
import torch
from torch import nn
from torch.nn import init # 初始化参数
import numpy as np
import sys
sys.path.append('.')
import d2lzh_pytorch
print(torch.__version__)
# 读取数据
batch_size = 256
train_iter, test_iter = d2lzh_pytorch.load_data_fashion_mnist(batch_size)
# 定义和初始化模型
num_inputs = 784
num_outputs = 10
# class LinearNet(nn.Module):
# def __init__(self, num_inputs, num_outputs):
# super(LinearNet, self).__init__()
# self.linear = nn.Linear(num_inputs, num_outputs)
# def forward(self, x): # x shape: (batch, 1, 28, 28)
# y = self.linear(x.view(x.shape[0], -1))
# return y
# net = LinearNet(num_inputs, num_outputs)
nn.ReLU(),
nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
)
num_channels, growth_rate = 64, 32
num_convs_in_dense_blocks = [4, 4, 4, 4]
for i, num_convs in enumerate(num_convs_in_dense_blocks):
DB = DenseBlock(num_convs, num_channels, growth_rate)
net.add_module("DenseBlock_%d" % i, DB)
num_channels = DB.out_channels
if i != len(num_convs_in_dense_blocks) - 1:
net.add_module("transition_block_%d" % i, transition_block(num_channels, num_channels // 2))
num_channels = num_channels // 2
net.add_module("BN", nn.BatchNorm2d(num_channels))
net.add_module("relu", nn.ReLU())
net.add_module("global_avg_pool", d2l.GlobalAvgPool2d())
net.add_module("fc", nn.Sequential(d2l.FlattenLayer(), nn.Linear(num_channels, 10)))
X = torch.rand(1, 1, 96, 96)
for name, layer in net.named_children():
X = layer(X)
print(name, 'output shape:\t', X.shape)
batch_size = 256
trian_iter, test_iter = d2l.load_data_fashion_mnist(batch_size, resize=96)
lr, num_epochs = 0.001, 5
optimizer = torch.optim.Adam(net.parameter(), lr=lr)
d2l.train_ch5(net, trian_iter, test_iter, batch_size, optimizer, device, num_epochs)
import torch
import torchvision
import torchvision.transforms as transforms
import matplotlib.pyplot as plt
import time
import sys
import d2lzh_pytorch as d2l
from PIL import Image
mnist_train = torchvision.datasets.FashionMNIST(
root="./DataSets/FashionMNIST",
train=True,
download=True,
transform=transforms.ToTensor())
mnist_test = torchvision.datasets.FashionMNIST(root="./DataSets/FashionMNIST",
train=False,
download=True,
transform=transforms.ToTensor())
X, y = [], []
for i in range(10):
X.append(mnist_test[i][0])
y.append(mnist_test[i][1])
d2l.show_fashion_mnist(X, d2l.get_fashion_mnist_labels(y))
train_iter, test_iter = d2l.load_data_fashion_mnist(256)
start = time.time()
for X, y in train_iter:
continue
print('%.2f sec' % (time.time() - start))
train_iter = torch.utils.data.DataLoader(mnist_train,
batch_size=batch_size,
shuffle=True,
num_workers=num_workers)
test_iter = torch.utils.data.DataLoader(mnist_test,
batch_size=batch_size,
shuffle=False,
num_workers=num_workers)
start = time.time()
for X, y in train_iter:
continue
print('%.2f sec' % (time.time() - start))
#####################softmax 从零开始实现#################################<<<
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(
batch_size, root='Data/softmax/FashionMNIST2065')
#模型参数初始化
num_inputs = 784
print(28 * 28)
num_outputs = 10
W = torch.tensor(np.random.normal(0, 0.01, (num_inputs, num_outputs)),
dtype=torch.float)
b = torch.zeros(num_outputs, dtype=torch.float)
W.requires_grad_(requires_grad=True)
b.requires_grad_(requires_grad=True)
#定义softmax
def softmax(X):
X_exp = X.exp()
partition = X_exp.sum(dim=1, keepdim=True)
# 模块5
# 2个Inception块接全局平均池化层
b5 = nn.Sequential(
Inception(832, 256, (160, 320), (32, 128), 128), # 256+320+128+128=832
Inception(832, 384, (192, 384), (48, 128), 128), # 384+384+128+128=1024
d2l.GlobalAvgPool2d())
# 总模型
# 接一个维度转换和全连接层
net = nn.Sequential(b1, b2, b3, b4, b5, d2l.FlattenLayer(),
nn.Linear(1024, 10))
# 测试
# GoogLeNet计算复杂，而且不如vgg那样便于修改通道数
# 将输入的高宽从224降到96
# x = torch.rand(1,1,96,96)
# for blk in net.children():
# x = blk(x)
# print(x.shape)
# 参数
lr, num_epochs, batch_size = 0.001, 5, 128
# 数据
train_iter, test_iter = d2l.load_data_fashion_mnist(
batch_size, resize=96) # 为减小数据规模，resize减小为96
# 优化器
optimizer = optim.Adam(net.parameters(), lr=lr)
# 训练
d2l.train_ch5(net, train_iter, test_iter, batch_size, optimizer, device,
num_epochs)