系列文章目录

机器学习笔记——梯度下降、反向传播
机器学习笔记——用pytorch实现线性回归
机器学习笔记——pytorch实现逻辑斯蒂回归Logistic regression
机器学习笔记——多层线性（回归）模型 Multilevel (Linear Regression) Model
深度学习笔记——pytorch构造数据集 Dataset and Dataloader
深度学习笔记——pytorch解决多分类问题 Multi-Class Classification
深度学习笔记——pytorch实现卷积神经网络CNN
深度学习笔记——卷积神经网络CNN进阶
深度学习笔记——循环神经网络 RNN
深度学习笔记——pytorch实现双向GRU

---

---

前言

参考视频——B站刘二大人《pytorch深度学习实践》

---

一、压缩填充张量 Pack_padded_sequence

在交给模型处理数据之前，我们需要将数据做成矩阵。
由于每条序列的长短不一，我们将一个batch_size的序列做成矩阵时，需要选取最长的序列作为矩阵的宽，在其他序列填充0，形成矩阵。
但我们在计算时这些0就是无用的数据，浪费计算资源
因此提出了压缩填充张量 Pack_padded_sequence

将矩阵转置后，并按序列长度排序

经过embedding层处理

记录有效数据，在计算时只计算有效数据

二、代码

代码如下（示例）：

#!/user/bin/env python3
# -*- coding: utf-8 -*-
import torch
from torch.utils.data import Dataset
from torch.utils.data import DataLoader
import pandas as pd
import time
import matplotlib.pyplot as plt
import math


BATCH_SIZE = 256  # batch size
HIDDEN_SIZE = 100  # 隐层维度
N_LAYER = 2  # RNN层数
N_EPOCHS = 100  # 训练轮数
N_CHARS = 128  # 字符
USE_GPU = True  # 是否使用gpu


# prepare data
class NameDataset(Dataset):
    def __init__(self, is_train_set=True):
        filename = 'data/names_train.csv' if is_train_set else 'data/names_test.csv'
        data = pd.read_csv(filename, delimiter=',', names=['names', 'country'])
        self.names = data['names']
        self.len = len(self.names)
        self.countries = data['country']
        self.countries_list = list(sorted(set(self.countries)))
        self.countries_dict = self.getCountryDict()
        self.countries_num = len(self.countries_list)

    def __getitem__(self, item):
        return self.names[item], self.countries_dict[self.countries[item]]

    def __len__(self):
        return self.len

    def getCountryDict(self):
        country_dict = {
    }
        for idx, country in enumerate(self.countries_list, 0):
            country_dict[country] = idx
        return country_dict

    def id2country(self, idx):
        return self.countries[idx]

    def getCountryNum(self):
        return self.countries_num



# 训练集
train_data = NameDataset(is_train_set=True)
trainloader = DataLoader(train_data, batch_size=BATCH_SIZE, shuffle=True)
# 测试集
test_data = NameDataset(is_train_set=False)
testloader = DataLoader(test_data, batch_size=BATCH_SIZE, shuffle=False)

N_COUNTRY = train_data.getCountryNum()  # 国家的数量


# 模型
class RNNClassifier(torch.nn.Module):
    def __init__(self, input_size, hidden_size, output_size, n_layer=1, bidirectional=True):
        super(RNNClassifier, self).__init__()
        self.hidden_size = hidden_size
        self.n_layer = n_layer
        self.n_directions = 2 if bidirectional else 1

        self.emb = torch.nn.Embedding(input_size, hidden_size)
        self.gru = torch.nn.GRU(hidden_size, hidden_size, num_layers=n_layer,
                                bidirectional=bidirectional)
        self.fc = torch.nn.Linear(hidden_size * self.n_directions, output_size)

    def forward(self, inputs, seq_lengths):
        inputs = create_tensor(inputs.t())
        batch_size = inputs.size(1)

        hidden = self._init_hidden(batch_size)
        embedding = self.emb(inputs)

        gru_input = torch.nn.utils.rnn.pack_padded_sequence(embedding, seq_lengths)  # 用于提速

        output, hidden = self.gru(gru_input, hidden)
        if self.n_directions == 2:
            # 如果是双向神经网络，则有两个hidden，需要将它们拼接起来
            hidden_cat = torch.cat([hidden[-1], hidden[-2]], dim=1)
        else:
            hidden_cat = hidden[-1]
        fc_output = self.fc(hidden_cat)
        return fc_output

    def _init_hidden(self, batch_size):
        hidden = torch.zeros(self.n_layer * self.n_directions, batch_size, self.hidden_size)
        return create_tensor(hidden)


def create_tensor(tensor):
    if USE_GPU:
        device = torch.device('cuda')
        tensor = tensor.to(device)
    return tensor


def make_tensors(names, countries):
    sequences_and_lengths = [name2list(name) for name in names]  # 得到name所有字符的ASCII码值和name的长度
    name_sequences = [sl[0] for sl in sequences_and_lengths]  # 获取name中所有字符的ASCII码值
    seq_lengths = torch.LongTensor([sl[1] for sl in sequences_and_lengths])  # 获取所有name的长度
    # 获得所有name的tensor，形状 batch_size*max(seq_len)  即name的个数*最长的name的长度
    seq_tensor = torch.zeros(len(name_sequences), seq_lengths.max()).long()  # 形状[name的个数*最长的name的长度]
    for idx, (seq, seq_len) in enumerate(zip(name_sequences, seq_lengths), 0):
        seq_tensor[idx, :seq_len] = torch.LongTensor(seq)  # 将所有name逐行填充到seq_tensor中

    #   sort by length to use pack_padded_sequence
    seq_lengths, perm_idx = seq_lengths.sort(dim=0, descending=True)  # 将seq_lengths按降序排列,perm_idx是排序后的序号
    seq_tensor = seq_tensor[perm_idx]  # seq_tensor中的顺序也随之改变
    countries = countries[perm_idx]  # countries中的顺序也随之改变

    # 返回所有names转为ASCII码的tensor，所有names的长度的tensor，所有country的tensor
    return create_tensor(seq_tensor), \
        create_tensor(seq_lengths), \
        create_tensor(countries)


def name2list(name):
    arr = [ord(c) for c in name]  # 将string转为list且所有字符转为ASCII码值
    return arr, len(arr)  # 返回的是tuple([arr],len(arr))


def modelTrain():
    total_loss = 0.0
    for i, (names, countries) in enumerate(trainloader, 1):
        inputs, seq_lengths, targets = make_tensors(names, countries)

        output = Net(inputs, seq_lengths.to('cpu'))
        loss = criterion(output, targets)
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        total_loss += loss.item()
        if i % 10 == 0:  # 每十个批次输出一次
            print(f'[{
      time_since(start_time)}] Epoch {
      epoch}', end='')
            print(f'[{
      i * len(inputs)}/{
      len(train_data)}]', end='')
            print(f'loss={
      total_loss / i * len(inputs)}')
    return total_loss  # 返回一轮训练的所有loss之和


def modelTest():
    correct = 0
    total = len(test_data)
    print('evaluating trained model...')
    with torch.no_grad():
        for i, (names, countries) in enumerate(testloader, 1):
            inputs, seq_lengths, targets = make_tensors(names, countries)
            output = Net(inputs, seq_lengths.to('cpu'))
            pred = output.max(dim=1, keepdim=True)[1]
            correct += pred.eq(targets.view_as(pred)).sum().item()

        percent = '%.2f' % (100 * correct / total)
        print(f'Test set:Accuracy{
      correct}/{
      total} {
      percent}%')
    return correct / total


def time_since(since):
    s = time.time() - since
    m = math.floor(s / 60)
    s -= m * 60
    return '%dm %ds' % (m, s)


if __name__ == '__main__':
    Net = RNNClassifier(N_CHARS, HIDDEN_SIZE, N_COUNTRY, N_LAYER, bidirectional=True)
    if USE_GPU:
        device = torch.device('cuda:0')
        Net.to(device)

    criterion = torch.nn.CrossEntropyLoss()
    optimizer = torch.optim.Adam(Net.parameters(), lr=0.001)

    start_time = time.time()
    print('Training for %d epochs...' % N_EPOCHS)
    acc_list = []
    epoch_list=[]
    for epoch in range(1, N_EPOCHS + 1):
        modelTrain()
        acc = modelTest()
        acc_list.append(acc)
        epoch_list.append(epoch)

    plt.plot(epoch_list, acc_list)
    plt.xlabel('epoch')
    plt.ylabel('accuracy')
    plt.grid()
    plt.show()

运行结果

---