在 中文 中使用 词向量 的示例及其翻译为 英语
{-}
-
Political
-
Ecclesiastic
-
Programming
词向量和word2vec.
语言建模概述词向量模型与语言模型非常紧密地交织在一起。
Word embedding models are quite closely intertwined with language models.Word2vec，是一群用来产生词向量的相关模型。
Word2vec is a group ofrelated models that are used to produce word embeddings.词向量模型与语言模型非常紧密地交织在一起。
Word embedding models are quite closely intertwined with language models.词向量方法是无监督式学习的少数几个成功应用之一。
Word vectors are considered to be among a number of successful applications of unsupervised learning.在寻找表示这些词向量的最优方法方面已经有许多进步。
There have been numerous advancements infinding the most optimal ways to represent these word vectors.例如，word2vec是一组浅层模型，用于生成词向量。
For example, word2vec is a group of shallowtwo-layer models that are used for producing word embeddings.好吧，那么现在我们有了我们的词向量，让我们看看它们如何与循环神经网络结合在一起的。
Okay, so now that we have our word vectors, let's see how they fit into recurrent neural networks.简介：我将讲解深度学习用于自然语言处理时的基础知识：词向量、循环神经网络、受语言学影响的任务和模型。
I will describe the foundations ofdeep learning for natural language processing: word vectors, recurrent neural networks, tasks and models influenced by linguistics.有人认为，Word2Vec最有趣的贡献是使得不同词向量之间表现出线性关系。
The most interesting contribution of Word2Vecwas the appearance of linear relationships between different word vectors.词向量表示法捕获许多语言属性，如性别，时态，复数甚至语义概念，如“首都城市”。
Word vector representations capture many linguistic properties such as gender, tense, plurality and even semantic concepts like"capital city of".首先，它们继承了词向量的一个重要特性：词的语义（semantics）。
First, they inherit an important property of the word vectors: 之前，这些NLP的无监督技术（例如GLoVe和word2vec）使用的都还是简单模型（词向量）和训练信号（词局部共现，thelocalco-occurenceofwords）。
Until recently, these unsupervised techniques for NLP(for example, GLoVe and word2vec)used simple models(word vectors) and training signals(the local co-occurence of 词向量出现在两个地方：.
The word cure appears in two places.词向量中出现的一些有趣的语义Pattern.
若需下载157种不同语言的预训练单词向量，可查看FastText。
To download pre-trained word vectors in 157 different languages, take a look at FastText.近年来，词向量已逐渐成为自然语言处理的基础知识。
Over the last few years, word embedding has gradually become basic knowledge in natural language processing.在这种情况下，单元只是新的词向量xt的函数。
In this scenario,先记住softmax的概念，因为后续许多词向量模型都会用到它。
Keep this softmax layer in mind, as many of the subsequent word embedding models will use it in some fashion.词向量方法是无监督式学习的少数几个成功应用之一。
Word embeddings are one of the few currently successful applications of unsupervised learning.从头开始训练CNN，不需要像word2vec或GloVe这样的预训练的单词向量。
Trains a CNN from scratch, without the need for for pre-trained word vectors like word2vec or GloVe.直到最近，这些无监督的NLP技术（例如GLoVe和word2vec）使用了简单的模型（词向量）和训练信号（单词的局部同时出现）。
Until recently, these unsupervised techniques for NLP(for example, GLoVe and word2vec)used simple models(word vectors) and training signals(the local co-occurence of Bengio等人在2003年首先提出了词向量的概念，当时是将其与语言模型的参数一并训练得到的。
The term word embeddings was originally coined by Bengio et al. in 2003 who trained them in a neural language model together with 通过一个非常简单的算法，我们可以获得丰富的词向量和段落向量，这些向量数据可以被应用到各种各样的NLP应用中。
With an extremely straightforward algorithm we can acquire rich word and paragraph vectors that could be utilized in a variety of NLP applications.此外，我们将投影矩阵W限制为正交矩阵，从而保持词嵌入向量之间的初始距离。
Additionally, we constrain the projector matrix W to be orthogonal so thatthe original distances between word embedding vectors are preserved.这样一来，给定一个相关的词，前述的向量空间就能用来计算最接近的前几个词。
词向量（WordVectors）.
Word(Thought) Vectors.从这句话，我们要为每个特定的词都创建一个词向量。
From this sentence, we want to create a word vector for each unique word.若是用100维的词向量表示一句10个单词的句子，我们将得到一个10x100维的矩阵作为输入。
For a 10 word sentence using a 100-dimensional embedding we would have a 10×100 matrix as our input.
