当前位置：网站首页>双向LSTM
双向LSTM
2022-08-04 05:29:00 【TigerZ*】
RNN和LSTM都只能依据之前时刻的时序信息来预测下一时刻的输出，但在有些问题中，当前时刻的输出不仅和之前的状态有关，还可能和未来的状态有关系。比如预测一句话中缺失的单词不仅需要根据前文来判断，还需要考虑它后面的内容，真正做到基于上下文判断。BRNN有两个RNN上下叠加在一起组成的，输出由这两个RNN的状态共同决定。
对于每个时刻t，输入会同时提供给两个方向相反的RNN，输出由这两个单向RNN共同决定。
参考链接：
边栏推荐
猜你喜欢
AIDL communication between two APPs
TensorFlow2 study notes: 6. Overfitting and underfitting, and their mitigation solutions
![[Deep Learning 21 Days Learning Challenge] Memo: What does our neural network model look like? - detailed explanation of model.summary()](/img/99/819ccbfed599ffd52307235309cdc9.png)
[Deep Learning 21 Days Learning Challenge] Memo: What does our neural network model look like? - detailed explanation of model.summary()
（十四）平衡二叉树
（九）哈希表
动手学深度学习_softmax回归
Dictionary feature extraction, text feature extraction.
【论文阅读】Further Non-local and Channel Attention Networks for Vehicle Re-identification
Pytorch语义分割理解
【论文阅读】Exploring Spatial Significance via Hybrid Pyramidal Graph Network for Vehicle Re-identificatio
随机推荐
【CV-Learning】卷积神经网络预备知识
YOLOV5 V6.1 详细训练方法
Deep Adversarial Decomposition: A Unified Framework for Separating Superimposed Images
安装dlib踩坑记录，报错：WARNING: pip is configured with locations that require TLS/SSL
read and study
JPEG2jpg
[CV-Learning] Semantic Segmentation
简单明了，数据库设计三大范式
with recursive用法
PP-LiteSeg
（十六）图的基本操作---两种遍历
（九）哈希表
[Deep Learning 21 Days Learning Challenge] 1. My handwriting was successfully recognized by the model - CNN implements mnist handwritten digit recognition model study notes
oracle的number与postgresql的numeric对比
动手学深度学习_卷积神经网络CNN
[CV-Learning] Convolutional Neural Network Preliminary Knowledge
YOLOV4流程图（方便理解）
【CV-Learning】目标检测＆实例分割
（十二）树--哈夫曼树
线性回归简介01---API使用案例