简体   繁体   English

如何替换 Tensorflow 中图形的现有节点?

[英]How to replace existing nodes of a graph in Tensorflow?

I already have a pretrained graph with me.I did some calculations on the existing weight values to get fused batchnorm weights(I have fused batchnorm and convolution weights).我已经有了一个预训练图。我对现有的权重值进行了一些计算以获得融合的batchnorm权重(我已经融合了batchnorm和卷积权重)。 Using my code, I was able to save these new weight to files.使用我的代码,我能够将这些新重量保存到文件中。 Now I wanted to replace this existing weights with new weights on the graph.现在我想用图表上的新权重替换现有的权重。 How this can be done?如何做到这一点? Any help would be appreciated.I am pretty new to this.任何帮助将不胜感激。我对此很陌生。

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.

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