Keras嵌入，“权重"在哪里?争论? [英] Keras Embedding ,where is the "weights" argument?

问题描述

我看到了如下代码:

embed_word = Embedding(params['word_voc_size'], params['embed_dim'], weights=[word_embed_matrix], input_length = params['word_max_size']
                        , trainable=False, mask_zero=True)

当我在Keras网站[ https://faroit.github.io/keras-docs/2.1.5/layers/embeddings/] [1]

When I look up the document in Keras website [https://faroit.github.io/keras-docs/2.1.5/layers/embeddings/][1]

我没有看到权重参数，

keras.layers.Embedding(input_dim, output_dim, embeddings_initializer='uniform', embeddings_regularizer=None, activity_regularizer=None, embeddings_constraint=None, mask_zero=False, input_length=None)

所以我很困惑，为什么我们可以使用Keras文档中未定义的参数权重?

So I am confused,why we can use the argument weights which was not defined the in Keras document?

我的keras版本是2.1.5.希望有人能帮助我.

My keras version is 2.1.5. Hope someone can help me.

推荐答案

Keras的Embedding层是Layer类的子类(每个Keras层都这样做). weights属性是在此基类中实现的，因此每个子类都将允许通过weights参数设置此属性.这也是为什么您不会在文档或Embedding层本身的实现中找到它的原因.

Keras' Embedding layer subclasses the Layer class (every Keras layer does this). The weights attribute is implemented in this base class, so every subclass will allow to set this attribute through a weights argument. This is also why you won't find it back in the documentation or the implementation of the Embedding layer itself.

您可以在这里(对于重量"，按Ctrl + F).

You can check the base layer implementation here (Ctrl + F for 'weight').

这篇关于Keras嵌入，“权重"在哪里?争论?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！