NNDL 实验七 循环神经网络（1）RNN记忆能力实验

循环神经网络（Recurrent Neural Network，RNN）是一类具有短期记忆能力的神经网络．

在循环神经网络中，神经元不但可以接受其他神经元的信息，也可以接受自身的信息，形成具有环路的网络结构．

和前馈神经网络相比，循环神经网络更加符合生物神经网络的结构．

目前，循环神经网络已经被广泛应用在语音识别、语言模型以及自然语言生成等任务上．

简单循环网络在参数学习时存在长程依赖问题，很难建模长时间间隔（Long Range）的状态之间的依赖关系。

为了测试简单循环网络的记忆能力，本节构建一个【数字求和任务】进行实验。

数字求和任务的输入是一串数字，前两个位置的数字为0-9，其余数字随机生成（主要为0），预测目标是输入序列中前两个数字的加和。图6.3展示了长度为10的数字序列．

6.1 循环神经网络的记忆能力实验
循环神经网络的一种简单实现是简单循环网络（Simple Recurrent Network，SRN）

简单循环网络在参数学习时存在长程依赖问题，很难建模长时间间隔（Long Range）的状态之间的依赖关系。为了测试简单循环网络的记忆能力，本节构建一个数字求和任务进行实验。

 

数字求和任务的输入是一串数字，前两个位置的数字为0-9，其余数字随机生成（主要为0），预测目标是输入序列中前两个数字的加和。图6.3展示了长度为10的数字序列．

 

如果序列长度越长，准确率越高，则说明网络的记忆能力越好．因此，我们可以构建不同长度的数据集，通过验证简单循环网络在不同长度的数据集上的表现，从而测试简单循环网络的长程依赖能力. 

6.1.1 数据集构建
构建不同长度的数字预测数据集DigitSum

6.1.1.1 数据集的构建函数

import os
import torch
import random
import numpy as np
import torch.nn as nn
 
# 固定随机种子
random.seed(0)
np.random.seed(0)
 
def generate_data(length, k, save_path):
    if length < 3:
        raise ValueError("The length of data should be greater than 2.")
    if k == 0:
        raise ValueError("k should be greater than 0.")
    # 生成100条长度为length的数字序列，除前两个字符外，序列其余数字暂用0填充
    base_examples = []
    for n1 in range(0, 10):
        for n2 in range(0, 10):
            seq = [n1, n2] + [0] * (length - 2)
            label = n1 + n2
            base_examples.append((seq, label))
 
    examples = []
    # 数据增强：对base_examples中的每条数据，默认生成k条数据，放入examples
    for base_example in base_examples:
        for _ in range(k):
            # 随机生成替换的元素位置和元素
            idx = np.random.randint(2, length)
            val = np.random.randint(0, 10)
            # 对序列中的对应零元素进行替换
            seq = base_example[0].copy()
            label = base_example[1]
            seq[idx] = val
            examples.append((seq, label))
 
    # 保存增强后的数据
    with open(save_path, "w", encoding="utf-8") as f:
        for example in examples:
            # 将数据转为字符串类型，方便保存
            seq = [str(e) for e in example[0]]
            label = str(example[1])
            line = " ".join(seq) + "\t" + label + "\n"
            f.write(line)
 
    print(f"generate data to: {save_path}.")
 
# 定义生成的数字序列长度
lengths = [5, 10, 15, 20, 25, 30, 35]
for length in lengths:
    # 首先判断是否存在这样的数据文件，不存在就建立一个
    if not os.path.exists(f"./datasets/{length}/"):
        os.makedirs(f"./datasets/{length}")
 
    # 生成长度为length的训练数据
    save_path = f"./datasets/{length}/train.txt"
    k = 3
    generate_data(length, k, save_path)
    # 生成长度为length的验证数据
    save_path = f"./datasets/{length}/dev.txt"
    k = 1
    generate_data(length, k, save_path)
    # 生成长度为length的测试数据
    save_path = f"./datasets/{length}/test.txt"
    k = 1
    generate_data(length, k, save_path)

6.1.1.2 加载数据并进行数据划分

# 加载数据
def load_data(data_path):
    # 加载训练集
    train_examples = []
    train_path = os.path.join(data_path, "train.txt")
    with open(train_path, "r", encoding="utf-8") as f:
        for line in f.readlines():
            # 解析一行数据，将其处理为数字序列seq和标签label
            items = line.strip().split("\t")
            seq = [int(i) for i in items[0].split(" ")]
            label = int(items[1])
            train_examples.append((seq, label))
 
    # 加载验证集
    dev_examples = []
    dev_path = os.path.join(data_path, "dev.txt")
    with open(dev_path, "r", encoding="utf-8") as f:
        for line in f.readlines():
            # 解析一行数据，将其处理为数字序列seq和标签label
            items = line.strip().split("\t")
            seq = [int(i) for i in items[0].split(" ")]
            label = int(items[1])
            dev_examples.append((seq, label))
 
    # 加载测试集
    test_examples = []
    test_path = os.path.join(data_path, "test.txt")
    with open(test_path, "r", encoding="utf-8") as f:
        for line in f.readlines():
            # 解析一行数据，将其处理为数字序列seq和标签label
            items = line.strip().split("\t")
            seq = [int(i) for i in items[0].split(" ")]
            label = int(items[1])
            test_examples.append((seq, label))
 
    return train_examples, dev_examples, test_examples
 
# 设定加载的数据集的长度
length = 5
# 该长度的数据集的存放目录
data_path = f"./datasets/{length}"
# 加载该数据集
train_examples, dev_examples, test_examples = load_data(data_path)
print("dev example:", dev_examples[:2])
print("训练集数量：", len(train_examples))
print("验证集数量：", len(dev_examples))
print("测试集数量：", len(test_examples))

 

6.1.1.3 构造Dataset类

class DigitSumDataset(Dataset):
    def __init__(self, data):
        self.data = data
 
    def __getitem__(self, idx):
        example = self.data[idx]
        seq = torch.tensor(example[0], dtype=torch.int64)
        label = torch.tensor(example[1], dtype=torch.int64)
        return seq, label
 
    def __len__(self):
        return len(self.data)

6.1.2 模型构建

6.1.2.1 嵌入层

class Embedding(nn.Module):
    def __init__(self, num_embeddings, embedding_dim):
        super(Embedding, self).__init__()
        # 定义嵌入矩阵
        self.W = torch.nn.Parameter(torch.Tensor(num_embeddings, embedding_dim))
 
    def forward(self, inputs):
        # 根据索引获取对应词向量
        embs = self.W[inputs]
        return embs
 
emb_layer = Embedding(10, 5)
inputs = torch.tensor([0, 1, 2, 3])
emb_layer(inputs)
print(emb_layer(inputs))

6.1.2.2 SRN层

自定义简单循环网络
飞桨框架内置了SRN的API paddle.nn.SimpleRNN 
将自己实现的SRN和Paddle框架内置的SRN返回的结果进行打印展示

import torch.nn.functional as F
torch.manual_seed(0)
 
# SRN模型
class SRN(nn.Module):
    def __init__(self, input_size,  hidden_size, W_attr=None, U_attr=None, b_attr=None):
        super(SRN, self).__init__()
        # 嵌入向量的维度
        self.input_size = input_size
        # 隐状态的维度
        self.hidden_size = hidden_size
        # 定义模型参数W，其shape为 input_size x hidden_size
        self.W = torch.nn.Parameter(torch.Tensor(input_size, hidden_size), torch.float32, attr=W_attr)
        # 定义模型参数U，其shape为hidden_size x hidden_size
        self.U = torch.nn.Parameter(torch.Tensor(hidden_size, hidden_size), torch.float32, attr=U_attr)
        # 定义模型参数b，其shape为 1 x hidden_size
        self.b = torch.nn.Parameter(torch.Tensor(1, hidden_size), torch.float32, attr=b_attr)
 
    # 初始化向量
    def init_state(self, batch_size):
        hidden_state = torch.zeros([batch_size, self.hidden_size], dtype=torch.float32)
        return hidden_state
 
    # 定义前向计算
    def forward(self, inputs, hidden_state=None):
        # inputs: 输入数据, 其shape为batch_size x seq_len x input_size
        batch_size, seq_len, input_size = inputs.shape
 
        # 初始化起始状态的隐向量, 其shape为 batch_size x hidden_size
        if hidden_state is None:
            hidden_state = self.init_state(batch_size)
 
        # 循环执行RNN计算
        for step in range(seq_len):
            # 获取当前时刻的输入数据step_input, 其shape为 batch_size x input_size
            step_input = inputs[:, step, :]
            # 获取当前时刻的隐状态向量hidden_state, 其shape为 batch_size x hidden_size
            hidden_state = F.tanh(torch.matmul(step_input, self.W) + torch.matmul(hidden_state, self.U) + self.b)
        return hidden_state

# 初始化参数并运行
from torch.nn.parameter import Parameter
 
m = torch.nn.Linear(2, 2)
W_attr = torch.normal([[0.1, 0.2], [0.1, 0.2]], size=(2, 2), requires_grad=True)
U_attr = torch.normal([[0.0, 0.1], [0.1, 0.0]], size=(2, 2), requires_grad=True)
b_attr = torch.normal([[0.1, 0.1]], size=(2, 2), requires_grad=True)
 
m.W_attr = Parameter(W_attr)
m.U_attr = Parameter(U_attr)
m.b_attr = Parameter(b_attr)
 
srn = SRN(2, 2, W_attr=m.W_attr, U_attr=m.U_attr, b_attr=m.b_attr)
 
inputs = torch.tensor([[[1, 0], [0, 2]]], dtype=torch.float32)
hidden_state = srn(inputs)
print("hidden_state", hidden_state)

hidden_state Tensor(shape=[1, 2], dtype=float32, place=Place(gpu:0),stop_gradient=False, [[0.34261486, 0.4845248]])

将自己实现的SRN与Paddle内置的SRN在输出值的精度上进行对比

# 这里创建一个随机数组作为测试数据，数据shape为batch_size x seq_len x input_size
batch_size, seq_len, input_size = 8, 20, 32
inputs = torch.randn([batch_size, seq_len, input_size])
 
# 设置模型的hidden_size
hidden_size = 32
paddle_srn = nn.SimpleRNN(input_size, hidden_size)
self_srn = SRN(input_size, hidden_size)
 
self_hidden_state = self_srn(inputs)
paddle_outputs, paddle_hidden_state = paddle_srn(inputs)
 
print("self_srn hidden_state: ", self_hidden_state.shape)
print("torch_srn outpus:", paddle_outputs.shape)
print("torch_srn hidden_state:", paddle_hidden_state.shape)

self_srn hidden_state:  [8, 32]
torch_srn outpus: [8, 20, 32]
torch_srn hidden_state: [1, 8, 32] 

在进行实验时，首先定义输入数据inputs，然后将该数据分别传入Paddle内置的SRN与自己实现的SRN模型中，最后通过对比两者的隐状态输出向量。

# 这里创建一个随机数组作为测试数据，数据shape为batch_size x seq_len x input_size
batch_size, seq_len, input_size, hidden_size = 2, 5, 10, 10
inputs = torch.randn([batch_size, seq_len, input_size])
 
# 设置模型的hidden_size
 
torch_srn = nn.RNN(input_size, hidden_size, bias=False)
 
# 获取torch_srn中的参数，并设置相应的paramAttr,用于初始化SRN
W_attr = torch_srn.weight_ih_l0.T
U_attr = torch_srn.weight_hh_l0.T
self_srn = SRN(input_size, hidden_size, W_attr=W_attr, U_attr=U_attr)
 
# 进行前向计算，获取隐状态向量，并打印展示
self_hidden_state = self_srn(inputs)
torch_outputs, torch_hidden_state = torch_srn(inputs)
print("torch SRN:\n", torch_hidden_state.detach().numpy().squeeze(0))
print("self SRN:\n", self_hidden_state.detach().numpy())

paddle SRN:
 [[ 0.3246606  -0.05465741 -0.3090897  -0.51604617 -0.11149617  0.4267313
   0.47200006 -0.06585315  0.85319966  0.18898569]
 [-0.4299355  -0.6067489  -0.59150505  0.30245274 -0.03939498  0.61462754
   0.4030218   0.49883503  0.02484456 -0.38516262]]
self SRN:
 [[ 0.32466057 -0.05465744 -0.3090897  -0.51604617 -0.11149605  0.4267313
   0.47200006 -0.06585318  0.85319966  0.18898569]
 [-0.42993543 -0.6067488  -0.59150493  0.3024528  -0.03939501  0.61462754
   0.40302184  0.49883503  0.02484456 -0.38516262]]

import time
 
# 这里创建一个随机数组作为测试数据，数据shape为batch_size x seq_len x input_size
batch_size, seq_len, input_size, hidden_size = 2, 5, 10, 10
inputs = torch.randn([batch_size, seq_len, input_size])
 
# 实例化模型
self_srn = SRN(input_size, hidden_size)
paddle_srn = nn.SimpleRNN(input_size, hidden_size)
 
# 计算自己实现的SRN运算速度
model_time = 0
for i in range(100):
    strat_time = time.time()
    out = self_srn(inputs)
    # 预热10次运算，不计入最终速度统计
    if i < 10:
        continue
    end_time = time.time()
    model_time += (end_time - strat_time)
avg_model_time = model_time / 90
print('self_srn speed:', avg_model_time, 's')
 
# 计算Paddle内置的SRN运算速度
model_time = 0
for i in range(100):
    strat_time = time.time()
    out = paddle_srn(inputs)
    # 预热10次运算，不计入最终速度统计
    if i < 10:
        continue
    end_time = time.time()
    model_time += (end_time - strat_time)
avg_model_time = model_time / 90
print('paddle_srn speed:', avg_model_time, 's')

self_srn speed: 0.0016264581504632564 s
paddle_srn speed: 0.0004503216208923415 s

6.1.2.3 线性层

线性层直接使用paddle.nn.Linear算子。

6.1.2.4 模型汇总

在定义了每一层的算子之后，我们定义一个数字求和模型Model_RNN4SeqClass，该模型会将嵌入层、SRN层和线性层进行组合，以实现数字求和的功能.

# 基于RNN实现数字预测的模型
class Model_RNN4SeqClass(nn.Module):
    def __init__(self, model, num_digits, input_size, hidden_size, num_classes):
        super(Model_RNN4SeqClass, self).__init__()
        # 传入实例化的RNN层，例如SRN
        self.rnn_model = model
        # 词典大小
        self.num_digits = num_digits
        # 嵌入向量的维度
        self.input_size = input_size
        # 定义Embedding层
        self.embedding = Embedding(num_digits, input_size)
        # 定义线性层
        self.linear = nn.Linear(hidden_size, num_classes)
 
    def forward(self, inputs):
        # 将数字序列映射为相应向量
        inputs_emb = self.embedding(inputs)
        # 调用RNN模型
        hidden_state = self.rnn_model(inputs_emb)
        # 使用最后一个时刻的状态进行数字预测
        logits = self.linear(hidden_state)
        return logits
 
# 实例化一个input_size为4， hidden_size为5的SRN
srn = SRN(4, 5)
# 基于srn实例化一个数字预测模型实例
model = Model_RNN4SeqClass(srn, 10, 4, 5, 19)
# 生成一个shape为 2 x 3 的批次数据
inputs = torch.tensor([[1, 2, 3], [2, 3, 4]])
# 进行模型前向预测
logits = model(inputs)
print(logits)

6.1.3 模型训练
6.1.3.1 训练指定长度的数字预测模型

import os
import random
import numpy as np
 
# 训练轮次
num_epochs = 500
# 学习率
lr = 0.001
# 输入数字的类别数
num_digits = 10
# 将数字映射为向量的维度
input_size = 32
# 隐状态向量的维度
hidden_size = 32
# 预测数字的类别数
num_classes = 19
# 批大小 
batch_size = 8
# 模型保存目录
save_dir = "./checkpoints"
 
# 通过指定length进行不同长度数据的实验
def train(length):
    print(f"\n====> Training SRN with data of length {length}.")
    # 固定随机种子
    np.random.seed(0)
    random.seed(0)
    torch.manual_seed(0)
 
    # 加载长度为length的数据
    data_path = f"./datasets/{length}"
    train_examples, dev_examples, test_examples = load_data(data_path)
    train_set, dev_set, test_set = DigitSumDataset(train_examples), DigitSumDataset(dev_examples), DigitSumDataset(test_examples)
    train_loader = torch.io.DataLoader(train_set, batch_size=batch_size)
    dev_loader = torch.io.DataLoader(dev_set, batch_size=batch_size)
    test_loader = torch.io.DataLoader(test_set, batch_size=batch_size)
    # 实例化模型
    base_model = SRN(input_size, hidden_size)
    model = Model_RNN4SeqClass(base_model, num_digits, input_size, hidden_size, num_classes) 
    # 指定优化器
    optimizer = torch.optim.Adam(lr, parameters=model.parameters()) 
    # 定义评价指标
    metric = Accuracy()
    # 定义损失函数
    loss_fn = nn.CrossEntropyLoss()
 
    # 基于以上组件，实例化Runner
    runner = RunnerV3(model, optimizer, loss_fn, metric)
 
    # 进行模型训练
    model_save_path = os.path.join(save_dir, f"best_srn_model_{length}.pdparams")
    runner.train(train_loader, dev_loader, num_epochs=num_epochs, eval_steps=100, log_steps=100, save_path=model_save_path)
 
    return runner

import torch
# 新增准确率计算函数
def accuracy(preds, labels):
    """
    输入：
        - preds：预测值，二分类时，shape=[N, 1]，N为样本数量，多分类时，shape=[N, C]，C为类别数量
        - labels：真实标签，shape=[N, 1]
    输出：
        - 准确率：shape=[1]
    """
    print(preds)
    # 判断是二分类任务还是多分类任务，preds.shape[1]=1时为二分类任务，preds.shape[1]>1时为多分类任务
    if preds.shape[1] == 1:
        # 二分类时，判断每个概率值是否大于0.5，当大于0.5时，类别为1，否则类别为0
        # 使用'torch.can_cast'将preds的数据类型转换为float32类型
        preds = torch.can_cast((preds>=0.5).dtype,to=torch.float32)
    else:
        # 多分类时，使用'torch.argmax'计算最大元素索引作为类别
        preds = torch.argmax(preds,dim=1)
        torch.can_cast(preds.dtype,torch.int32)
    return torch.mean(torch.tensor((preds == labels), dtype=torch.float32))
 
 
class Accuracy():
    def __init__(self):
        """
        输入：
           - is_logist: outputs是logist还是激活后的值
        """
 
        # 用于统计正确的样本个数
        self.num_correct = 0
        # 用于统计样本的总数
        self.num_count = 0
 
        self.is_logist = True
 
    def update(self, outputs, labels):
        """
        输入：
           - outputs: 预测值, shape=[N,class_num]
           - labels: 标签值, shape=[N,1]
        """
 
        # 判断是二分类任务还是多分类任务，shape[1]=1时为二分类任务，shape[1]>1时为多分类任务
        if outputs.shape[1] == 1: # 二分类
            outputs = torch.squeeze(outputs, axis=-1)
            if self.is_logist:
                # logist判断是否大于0
                preds = torch.can_cast((outputs>=0), dtype=torch.float32)
            else:
                # 如果不是logist，判断每个概率值是否大于0.5，当大于0.5时，类别为1，否则类别为0
                preds = torch.can_cast((outputs>=0.5), dtype=torch.float32)
        else:
            # 多分类时，使用'paddle.argmax'计算最大元素索引作为类别
            preds = torch.argmax(outputs, dim=1).int()
 
        # 获取本批数据中预测正确的样本个数
        labels = torch.squeeze(labels, dim=-1)
        batch_correct = torch.sum(torch.tensor(preds == labels, dtype=torch.float32)).cpu().numpy()
        batch_count = len(labels)
 
        # 更新num_correct 和 num_count
        self.num_correct += batch_correct
        self.num_count += batch_count
 
    def accumulate(self):
        # 使用累计的数据，计算总的指标
        if self.num_count == 0:
            return 0
        return self.num_correct / self.num_count
 
    def reset(self):
        # 重置正确的数目和总数
        self.num_correct = 0
        self.num_count = 0
 
    def name(self):
        return "Accuracy"

class RunnerV3(object):
    def __init__(self, model, optimizer, loss_fn, metric, **kwargs):
        self.model = model
        self.optimizer = optimizer
        self.loss_fn = loss_fn
        self.metric = metric  # 只用于计算评价指标
 
        # 记录训练过程中的评价指标变化情况
        self.dev_scores = []
 
        # 记录训练过程中的损失函数变化情况
        self.train_epoch_losses = []  # 一个epoch记录一次loss
        self.train_step_losses = []  # 一个step记录一次loss
        self.dev_losses = []
 
        # 记录全局最优指标
        self.best_score = 0
 
    def train(self, train_loader, dev_loader=None, **kwargs):
        # 将模型切换为训练模式
        self.model.train()
 
        # 传入训练轮数，如果没有传入值则默认为0
        num_epochs = kwargs.get("num_epochs", 0)
        # 传入log打印频率，如果没有传入值则默认为100
        log_steps = kwargs.get("log_steps", 100)
        # 评价频率
        eval_steps = kwargs.get("eval_steps", 0)
 
        # 传入模型保存路径，如果没有传入值则默认为"best_model.pdparams"
        save_path = kwargs.get("save_path", "best_model.pdparams")
 
        custom_print_log = kwargs.get("custom_print_log", None)
 
        # 训练总的步数
        num_training_steps = num_epochs * len(train_loader)
 
        if eval_steps:
            if self.metric is None:
                raise RuntimeError('Error: Metric can not be None!')
            if dev_loader is None:
                raise RuntimeError('Error: dev_loader can not be None!')
 
        # 运行的step数目
        global_step = 0
 
        # 进行num_epochs轮训练
        for epoch in range(num_epochs):
            # 用于统计训练集的损失
            total_loss = 0
            for step, data in enumerate(train_loader):
                X, y = data
                X = X.cuda()
                y = y.cuda()
                # 获取模型预测
                logits = self.model(X)
                logits = logits.cuda()
                y = y.to(dtype=torch.int64)
                loss = self.loss_fn(logits, y)  # 默认求mean
                total_loss += loss
 
                # 训练过程中，每个step的loss进行保存
                self.train_step_losses.append((global_step, loss.item()))
 
                if log_steps and global_step % log_steps == 0:
                    print(
                        f"[Train] epoch: {epoch}/{num_epochs}, step: {global_step}/{num_training_steps}, loss: {loss.item():.5f}")
 
                # 梯度反向传播，计算每个参数的梯度值
                loss.backward()
 
                if custom_print_log:
                    custom_print_log(self)
 
                # 小批量梯度下降进行参数更新
                self.optimizer.step()
                # 梯度归零
                optimizer.zero_grad()
 
                # 判断是否需要评价
                if eval_steps > 0 and global_step > 0 and \
                        (global_step % eval_steps == 0 or global_step == (num_training_steps - 1)):
 
                    dev_score, dev_loss = self.evaluate(dev_loader, global_step=global_step)
                    print(f"[Evaluate]  dev score: {dev_score:.5f}, dev loss: {dev_loss:.5f}")
 
                    # 将模型切换为训练模式
                    self.model.train()
 
                    # 如果当前指标为最优指标，保存该模型
                    if dev_score > self.best_score:
                        self.save_model(save_path)
                        print(
                            f"[Evaluate] best accuracy performence has been updated: {self.best_score:.5f} --> {dev_score:.5f}")
                        self.best_score = dev_score
 
                global_step += 1
 
            # 当前epoch 训练loss累计值
            trn_loss = (total_loss / len(train_loader)).item()
            # epoch粒度的训练loss保存
            self.train_epoch_losses.append(trn_loss)
 
        print("[Train] Training done!")
 
    # 模型评估阶段，使用'paddle.no_grad()'控制不计算和存储梯度
    @torch.no_grad()
    def evaluate(self, dev_loader, **kwargs):
        assert self.metric is not None
 
        # 将模型设置为评估模式
        self.model.eval()
 
        global_step = kwargs.get("global_step", -1)
 
        # 用于统计训练集的损失
        total_loss = 0
 
        # 重置评价
        self.metric.reset()
 
        # 遍历验证集每个批次
        for batch_id, data in enumerate(dev_loader):
            X, y = data
            X = X.cuda()
            y = y.cuda()
            # 计算模型输出
            logits = self.model(X)
            logits = logits.cuda()
            # 计算损失函数
            y=y.to(dtype=torch.int64)
            loss = self.loss_fn(logits, y).item()
            # 累积损失
            total_loss += loss
 
            # 累积评价
            self.metric.update(logits, y)
 
        dev_loss = (total_loss / len(dev_loader))
        dev_score = self.metric.accumulate()
 
        # 记录验证集loss
        if global_step != -1:
            self.dev_losses.append((global_step, dev_loss))
            self.dev_scores.append(dev_score)
 
        return dev_score, dev_loss
 
    # 模型评估阶段，使用'paddle.no_grad()'控制不计算和存储梯度
    @torch.no_grad()
    def predict(self, x, **kwargs):
        # 将模型设置为评估模式
        self.model.eval()
        # 运行模型前向计算，得到预测值
        logits = self.model(x)
        return logits
 
    def save_model(self, save_path):
        torch.save(self.model.state_dict(), save_path)
 
    def load_model(self, model_path):
        state_dict = torch.load(model_path)
        self.model.load_state_dict(state_dict)

6.1.3.2 多组训练

srn_runners = {}
 
lengths = [10, 15, 20, 25, 30, 35]
for length in lengths:
    runner = train(length)
    srn_runners[length] = runner

6.1.3.3 损失曲线展示

图6.6展示了在6个数据集上的损失变化情况，数据集的长度分别为10、15、20、25、30和35. 从输出结果看，随着数据序列长度的增加，虽然训练集损失逐渐逼近于0，但是验证集损失整体趋向越来越大，这表明当序列变长时，SRN模型保持序列长期依赖能力在逐渐变弱，越来越无法学习到有用的知识.

 

import matplotlib.pyplot as plt
 
def plot_training_loss(runner, fig_name, sample_step):
    plt.figure()
    train_items = runner.train_step_losses[::sample_step]
    train_steps = [x[0] for x in train_items]
    train_losses = [x[1] for x in train_items]
    plt.plot(train_steps, train_losses, color='#e4007f', label="Train loss")
 
    dev_steps = [x[0] for x in runner.dev_losses]
    dev_losses = [x[1] for x in runner.dev_losses]
    plt.plot(dev_steps, dev_losses, color='#f19ec2', linestyle='--', label="Dev loss")
 
    # 绘制坐标轴和图例
    plt.ylabel("loss", fontsize='large')
    plt.xlabel("step", fontsize='large')
    plt.legend(loc='upper right', fontsize='x-large')
 
    plt.savefig(fig_name)
    plt.show()

# 画出训练过程中的损失图
for length in lengths:
    runner = srn_runners[length]
    fig_name = f"./images/6.6_{length}.pdf"
    plot_training_loss(runner, fig_name, sample_step=100)

6.1.4 模型评价

srn_dev_scores = []
srn_test_scores = []
for length in lengths:
    print(f"Evaluate SRN with data length {length}.")
    runner = srn_runners[length]
    # 加载训练过程中效果最好的模型
    model_path = os.path.join(save_dir, f"best_srn_model_{length}.pdparams")
    runner.load_model(model_path)
 
    # 加载长度为length的数据
    data_path = f"./datasets/{length}"
    train_examples, dev_examples, test_examples = load_data(data_path)
    test_set = DigitSumDataset(test_examples)
    test_loader = torch.io.DataLoader(test_set, batch_size=batch_size)
 
    # 使用测试集评价模型，获取测试集上的预测准确率
    score, _ = runner.evaluate(test_loader)
    srn_test_scores.append(score)
    srn_dev_scores.append(max(runner.dev_scores))
 
for length, dev_score, test_score in zip(lengths, srn_dev_scores, srn_test_scores):
    print(f"[SRN] length:{length}, dev_score: {dev_score}, test_score: {test_score: .5f}")

import matplotlib.pyplot as plt
 
plt.plot(lengths, srn_dev_scores, '-o', color='#e4007f',  label="Dev Accuracy")
plt.plot(lengths, srn_test_scores,'-o', color='#f19ec2', label="Test Accuracy")
 
#绘制坐标轴和图例
plt.ylabel("accuracy", fontsize='large')
plt.xlabel("sequence length", fontsize='large')
plt.legend(loc='upper right', fontsize='x-large')
 
fig_name = "./images/6.7.pdf"
plt.savefig(fig_name)
plt.show()

总结：通过这次实验，对循环神经网络更加了解，熟悉。