pytorch学习-- RNN, RNNCell

文章目录

• 循环神经网络
• RNN Layer
• 多层RNN
• nn.RNNCell
• 梯度爆炸
• 梯度离散

循环神经网络

$h_t 来自h_{t-1}与x_t$

求解导数

RNN Layer

$[batch, featureLen]@[hiddenLen, featureLen]^T$
$[batch, hiddenLen]@[hidden Len, hiddenLen]^T$
对应
$x_t@W_{xh} + h_t@W_{hh}$
$x:[seqLen,batch,featureLen]$ [10,3,100]
$x_t:[batch,featureLen]$ [3,100],3批次，一个单词

rnn = nn.RNN(100,10)	#词向量，隐藏层向量
rnn.weight_hh_l0.shape	#一个RNN有四个参数
rnn.bias_ih_l0.shape

三个参数，input_size，hidden_size, num_layers
可以通过设置batch_first参数设置维度表示方式

out,ht = forward(x,h0)
#注意此处x即为[seq,batch,wordvec]，直接将一句话喂进去
h0/ht -> [num_layers,batch,h_dim]
out -> [seq,batch,h_dim],他返回每个时间步的out

rnn = nn.RNN(100,10)	#词向量，隐藏层向量
x = torch.randn(10,3,100)
out,h = rnn(x,torch.zeros(1,3,20)

多层RNN

注意上层的$W_{ih}$的尺度变为【hidden，hidden】

nn.RNNCell

更加灵活，手动喂多次

三个参数，input_size，hidden_size, num_layers 初始化相同

ht = rnncell(xt,ht_1)
# xt:[b,wordvec]
# ht_1/ht:[num_layers,b,h_dim]
#out = torch.stack([h1,h2...ht]), h的集合

cell1 = nn.RNNCell(100,20)
h1 = torch.zeros(3,20)
for xt in x:
	h1 = cell1(xt,h1)

两层并且hidden不同

cell1 = nn.RNNCell(100,30)
cell2 = nn.RNNCell(30,20)
h1 = torch.zeros(3,30)
h2 = torch.zeros(3,20)
for xt in x:
	h1 = cell1(xt,h1)
	h2 = cell2(h1,h2)

梯度爆炸

对w.grad 进行clipping

loss = criterion(output,y)
model.zero_grad()
loss.backward()
for p in model.parameters():
	print(p.grad.norm())	#检测是否爆炸
	torch.nn.utils.clip_grad_norm_(p,10)
optimizer.step()

梯度离散

来源：CSDN

作者：bit_codertoo

链接：https://blog.csdn.net/bit_codertoo/article/details/103653652