RNN在文本預(yù)測等方面有著諸多使用，也是重要的神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)，其結(jié)構(gòu)包括RNN，LSTM，GRU等。本文以分類這一任務(wù)為基礎(chǔ)分析RNN的簡單結(jié)構(gòu)

1. 引入包和設(shè)置超參數(shù)

import torch
from torch import nn
import torchvision.datasets as dsets
import torchvision.transforms as transforms
import matplotlib.pyplot as plt


# torch.manual_seed(1)    # reproducible

# Hyper Parameters
EPOCH = 1               # train the training data n times, to save time, we just train 1 epoch
B_S = 64
TIME_STEP = 28          # rnn time step / image height
INPUT_SIZE = 28         # rnn input size / image width
LR = 0.01               # learning rate
DOWNLOAD_MNIST = True   # set to True if haven't download the data

2. 準(zhǔn)備MNIST數(shù)據(jù)

# Mnist digital dataset
train_data = dsets.MNIST(
    root='./mnist/',
    train=True,                         # this is training data
    transform=transforms.ToTensor(),    # Converts a PIL.Image or numpy.ndarray to
    # torch.FloatTensor of shape (C x H x W) and normalize in the range [0.0, 1.0]
    download=DOWNLOAD_MNIST,            # download it if you don't have it
)

# plot one example
print(train_data.train_data.size())     # (60000, 28, 28)
print(train_data.train_labels.size())   # (60000)
plt.imshow(train_data.train_data[0].numpy(), cmap='gray')
plt.title('%i' % train_data.train_labels[0])
plt.show()

# Data Loader for easy mini-batch return in training
train_loader = torch.utils.data.DataLoader(dataset=train_data, batch_size=B_S, shuffle=True)

# convert test data into Variable, pick 2000 samples to speed up testing
test_data = dsets.MNIST(root='./mnist/', train=False, transform=transforms.ToTensor())
test_x = test_data.test_data.type(torch.FloatTensor)[:2000]/255.   
# shape (2000, 28, 28) value in range(0,1)
test_y = test_data.test_labels.numpy()[:2000]    # covert to numpy array

注：

1. torchvision.datasets.MNIST(root, train, transform,download)：root為MNIST數(shù)據(jù)所在路徑，train為設(shè)置該數(shù)據(jù)集為訓(xùn)練數(shù)據(jù)(True)或檢驗(yàn)數(shù)據(jù)(False)，transform表示轉(zhuǎn)換數(shù)據(jù)至某種形式，download為設(shè)置是否下載該數(shù)據(jù)(若已下載則設(shè)置為False)
2. torch.utils.data.DataLoader(dataset, batch_size, shuffle)：裝載訓(xùn)練數(shù)據(jù)，其中dataset用于指定所裝載數(shù)據(jù)集路徑

3. 構(gòu)建RNN網(wǎng)絡(luò)

class RNN(nn.Module):
    def __init__(self):
        super(RNN, self).__init__()

        self.rnn = nn.LSTM(         # if use nn.RNN(), it hardly learns
            input_size=INPUT_SIZE,
            hidden_size=64,         # rnn hidden unit
            num_layers=1,           # number of rnn layer
            batch_first=True,       
# input & output will has batch size as 1s dimension. e.g. (batch, time_step, input_size)
        )

        self.out = nn.Linear(64, 10)

    def forward(self, x):
        # x shape (batch, time_step, input_size)
        # r_out shape (batch, time_step, output_size)
        # h_n shape (n_layers, batch, hidden_size)
        # h_c shape (n_layers, batch, hidden_size)
        r_out, (h_n, h_c) = self.rnn(x, None)   # None represents zero initial hidden state

        # choose r_out at the last time step
        out = self.out(r_out[:, -1, :])
        return out


rnn = RNN()
print(rnn)

注1：

1. self.rnn=nn.LSTM(input_size,hidden_size,num_layers )：設(shè)定第一層級的RNN網(wǎng)絡(luò)為LSTM，其中input_size表示輸入數(shù)據(jù)的維度，hidden_size表示隱藏神經(jīng)元數(shù)量，num_layers表示LSTM網(wǎng)絡(luò)的層數(shù)
2. self.out = nn.Linear(P1, P2)：P1表示隱藏神經(jīng)元個數(shù)，P2表示輸出類別數(shù)
3. LSTM存在兩個輸出和輸入，表示預(yù)測內(nèi)容與狀態(tài)

注2：不同LSTM比較

單層三隱藏神經(jīng)元LSTM

三層六隱藏神經(jīng)元LSTM

4. 選擇優(yōu)化器與損失函數(shù)

optimizer = torch.optim.Adam(rnn.parameters(), lr=LR)   # optimize all cnn parameters
loss_func = nn.CrossEntropyLoss()                       # the target label is not one-hotted

注：

1. adam()：最常用優(yōu)化器之一，為AdaGrad與動量算法的組合
2. nn.CrossEntropyLoss() ：判斷實(shí)際輸出與期望輸出的差異程度，多用于分類問題中判斷預(yù)測目標(biāo)概率分布與實(shí)際概率分布的差異

5. 訓(xùn)練和測試

# training and testing
for epoch in range(EPOCH):
    for step, (b_x, b_y) in enumerate(train_loader):        # gives batch data
        b_x = b_x.view(-1, 28, 28)              # reshape x to (batch, time_step, input_size)

        output = rnn(b_x)                               # rnn output
        loss = loss_func(output, b_y)                   # cross entropy loss
        optimizer.zero_grad()                           # clear gradients for this training step
        loss.backward()                                 # backpropagation, compute gradients
        optimizer.step()                                # apply gradients

        if step % 50 == 0:
            test_output = rnn(test_x)                   # (samples, time_step, input_size)
            pred_y = torch.max(test_output, 1)[1].data.numpy()
            accuracy = float((pred_y == test_y).astype(int).sum()) / float(test_y.size)
            print('Epoch: ', epoch, '| train loss: %.4f' % loss.data.numpy(), '| test accuracy: %.2f' % accuracy)

# print 10 predictions from test data
test_output = rnn(test_x[:10].view(-1, 28, 28))
pred_y = torch.max(test_output, 1)[1].data.numpy()
print(pred_y, 'prediction number')
print(test_y[:10], 'real number')