简体繁体 English

CNN 对 RNN 模型的权重

[英]CNN weights on RNN model

原文 2020-03-01 14:43:58 8 1 python

是否有可能在 RNN 模型上使用 CNN 权重，就像您可以使用 CNN 的权重来学习每个字母的样子并使用来自该 CNN 的权重来翻译整个句子吗？

1 个解决方案

It is possible.有可能的。 For an RNN you have to define what is the hidden state so you may set it to be the output of the CNN on the previous element in the sequence.对于 RNN，您必须定义隐藏状态是什么，以便您可以将其设置为 CNN 在序列中的前一个元素上的输出。 There are several models with state-of-the-art performance on sentence translation.有几种模型在句子翻译方面具有最先进的性能。 One of them is the Transformer which is makes use of convolutions, as in CNNs, and self-attention.其中之一是 Transformer，它利用卷积（如 CNN）和自注意力。 However, it doesn't do so in an RNN scheme, as RNNs are slower to train and evaluate due to their sequential nature.然而，它在 RNN 方案中不会这样做，因为 RNN 的训练和评估速度较慢，因为它们是顺序性的。 Also note that RNNs are inferior (mainly) due to their problem of short-term memory.另请注意，RNN 较差（主要是）由于它们的短期记忆问题。 You may want to read on LSTMs (that became their successor).您可能想阅读 LSTM（成为它们的继任者）。

观察张量流rnn模型权重 - Observing tensorflow rnn model weights

如何在Tensorflow CNN模型中初始化权重？ - How to initialize weights in tensorflow CNN model?

修改 TensorFlow 中恢复的 CNN model 的权重和偏差 - Modifying the weights and biases of a restored CNN model in TensorFlow

有没有办法训练一个 CNN 模型，保存这个 CNN 的权重，然后使用这个权重来为其他训练数据重新训练这个 CNN？ - Is there a way to train a CNN model, save the weights of this CNN, and then use this weights to retrain this CNN for other train data?

如何将RNN与CNN结合？ - How to combine RNN with CNN?

将 CNN 连接到 RNN - Connecting CNN to RNN

CNN 模型的权重变为非常小的值，并且在 NaN 之后 - Weights of CNN model go to really small values and after NaN

将RNN和CNN与烤宽面条结合 - combining RNN and CNN with lasagne

不平衡 CNN 上的类权重 - Class weights on imbalanced CNN

CNN 中的 Class 权重 - Class weights in CNN

暂无

暂无

声明:本站的技术帖子网页，遵循CC BY-SA 4.0协议，如果您需要转载，请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

相关问题 观察张量流rnn模型权重 - Observing tensorflow rnn model weights 如何在Tensorflow CNN模型中初始化权重？ - How to initialize weights in tensorflow CNN model? 修改 TensorFlow 中恢复的 CNN model 的权重和偏差 - Modifying the weights and biases of a restored CNN model in TensorFlow 有没有办法训练一个 CNN 模型，保存这个 CNN 的权重，然后使用这个权重来为其他训练数据重新训练这个 CNN？ - Is there a way to train a CNN model, save the weights of this CNN, and then use this weights to retrain this CNN for other train data? 如何将RNN与CNN结合？ - How to combine RNN with CNN? 将 CNN 连接到 RNN - Connecting CNN to RNN CNN 模型的权重变为非常小的值，并且在 NaN 之后 - Weights of CNN model go to really small values and after NaN 将RNN和CNN与烤宽面条结合 - combining RNN and CNN with lasagne 不平衡 CNN 上的类权重 - Class weights on imbalanced CNN CNN 中的 Class 权重 - Class weights in CNN

相关标签

粤ICP备18138465号 © 2020-2024 STACKOOM.COM