Tensorflow 2.0:冻结图支持 [英] Tensorflow 2.0 : frozen graph support

问题描述

tensorflow 2.0 会继续支持冻结图还是弃用?我的意思是从saved_model 创建/优化冻结图的脚本和API.还有用于运行推理的 API.

Will the support for frozen graph continue in tensorflow 2.0 or deprecated? I mean the scripts and APIs to create/optimize frozen graph from saved_model. Also the APIs to run the inference for the same.

假设将来会支持它，在 tensorflow 2.0 中对冻结图运行推理的推荐方法是什么?

Assuming it will be supported in future, what is the recommended method to run the inference on frozen graph in tensorflow 2.0 ?

推荐答案

冻结图 API - freeze_graph.py 和 converter_variables_to_constants - TensorFlow 2.0 不支持.

The freeze graph APIs - freeze_graph.py and converter_variables_to_constants - will not be supported in TensorFlow 2.0.

在 2.0 中，主要导出格式是 SavedModels，因此构建的 API 可以直接支持保存的模型.可以使用 v1.compat 路径对现有冻结图进行推理.

In 2.0, the primary export format is SavedModels so APIs are built to directly support SavedModels. Inference on existing frozen graphs can be run using the v1.compat path.

这篇关于Tensorflow 2.0:冻结图支持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！