TensorFlow 将图形保存到文件中/从文件加载图形 [英] TensorFlow saving into/loading a graph from a file

问题描述

从我目前收集到的信息来看，有几种不同的方法可以将 TensorFlow 图转储到文件中，然后将其加载到另一个程序中，但我无法找到有关它们如何工作的明确示例/信息.我已经知道的是:

From what I've gathered so far, there are several different ways of dumping a TensorFlow graph into a file and then loading it into another program, but I haven't been able to find clear examples/information on how they work. What I already know is this:

1. 使用 tf.train.Saver() 将模型的变量保存到检查点文件 (.ckpt) 中，并在稍后恢复它们 (source)
2. 将模型保存到 .pb 文件中，然后使用 tf.train.write_graph() 和 tf.import_graph_def()(源)
3. 从 .pb 文件加载模型，重新训练，然后使用 Bazel 将其转储到新的 .pb 文件中 (来源)
4. 冻结图形以将图形和权重保存在一起(来源)
5. 使用 as_graph_def() 保存模型，对于权重/变量，将它们映射到常量 (源)

1. Save the model's variables into a checkpoint file (.ckpt) using a tf.train.Saver() and restore them later (source)
2. Save a model into a .pb file and load it back in using tf.train.write_graph() and tf.import_graph_def() (source)
3. Load in a model from a .pb file, retrain it, and dump it into a new .pb file using Bazel (source)
4. Freeze the graph to save the graph and weights together (source)
5. Use as_graph_def() to save the model, and for weights/variables, map them into constants (source)

但是，关于这些不同的方法，我一直无法澄清几个问题:

However, I haven't been able to clear up several questions regarding these different methods:

1. 关于检查点文件，它们是否只保存模型的训练权重?检查点文件是否可以加载到新程序中，并用于运行模型，或者它们只是作为在特定时间/阶段保存模型中权重的方法?
2. 关于tf.train.write_graph()，权重/变量是否也保存了?
3. 关于 Bazel，它只能保存到 .pb 文件/从 .pb 文件加载以进行再训练吗?是否有一个简单的 Bazel 命令可以将图形转储到 .pb 中?
4. 关于冻结，可以使用tf.import_graph_def()加载冻结图吗?
5. TensorFlow 的 Android 演示从 .pb 文件加载到 Google 的 Inception 模型中.如果我想替换我自己的 .pb 文件，我该怎么做?我需要更改任何本机代码/方法吗?
6. 一般来说，所有这些方法之间究竟有什么区别?或者更广泛地说，as_graph_def()/.ckpt/.pb 之间有什么区别?

1. Regarding checkpoint files, do they only save the trained weights of a model? Could checkpoint files be loaded into a new program, and be used to run the model, or do they simply serve as ways to save the weights in a model at a certain time/stage?
2. Regarding tf.train.write_graph(), are the weights/variables saved as well?
3. Regarding Bazel, can it only save into/load from .pb files for retraining? Is there a simple Bazel command just to dump a graph into a .pb?
4. Regarding freezing, can a frozen graph be loaded in using tf.import_graph_def()?
5. The Android demo for TensorFlow loads in Google's Inception model from a .pb file. If I wanted to substitute my own .pb file, how would I go about doing that? Would I need to change any native code/methods?
6. In general, what exactly is the difference between all these methods? Or more broadly, what is the difference between as_graph_def()/.ckpt/.pb?

简而言之，我正在寻找一种将图形(例如各种操作等)及其权重/变量保存到文件中的方法，然后可以使用该文件加载图形和权重进入另一个程序，以供使用(不一定要继续/再培训).

In short, what I'm looking for is a method to save both a graph (as in, the various operations and such) and its weights/variables into a file, which can then be used to load the graph and weights into another program, for use (not necessarily continuing/retraining).

有关此主题的文档不是很简单，因此非常感谢您提供任何答案/信息.

Documentation about this topic isn't very straightforward, so any answers/information would be greatly appreciated.

推荐答案

有很多方法可以解决在 TensorFlow 中保存模型的问题，这可能会让人有点困惑.依次回答您的每个子问题:

There are many ways to approach the problem of saving a model in TensorFlow, which can make it a bit confusing. Taking each of your sub-questions in turn:

1. 检查点文件(例如通过调用saver.save() 在 tf.train.Saver 对象)仅包含权重，以及在同一程序中定义的任何其他变量.要在另一个程序中使用它们，您必须重新创建关联的图结构(例如，通过运行代码再次构建它，或调用 tf.import_graph_def())，它告诉 TensorFlow 如何处理这些权重.请注意，调用 saver.save() 还会生成一个包含 MetaGraphDef，其中包含一个图表以及如何将来自检查点的权重与该图表相关联的详细信息.有关详细信息，请参阅教程.

1. The checkpoint files (produced e.g. by calling saver.save() on a tf.train.Saver object) contain only the weights, and any other variables defined in the same program. To use them in another program, you must re-create the associated graph structure (e.g. by running code to build it again, or calling tf.import_graph_def()), which tells TensorFlow what to do with those weights. Note that calling saver.save() also produces a file containing a MetaGraphDef, which contains a graph and details of how to associate the weights from a checkpoint with that graph. See the tutorial for more details.

tf.train.write_graph() 只写图结构；不是权重.

tf.train.write_graph() only writes the graph structure; not the weights.

Bazel 与读取或写入 TensorFlow 图无关.(也许我误解了您的问题:请随时在评论中澄清.)

Bazel is unrelated to reading or writing TensorFlow graphs. (Perhaps I misunderstand your question: feel free to clarify it in a comment.)

可以使用 tf.import_graph_def() 加载冻结图.在这种情况下，权重(通常)嵌入在图中，因此您无需加载单独的检查点.

A frozen graph can be loaded using tf.import_graph_def(). In this case, the weights are (typically) embedded in the graph, so you don't need to load a separate checkpoint.

主要的变化是更新输入模型的张量的名称，以及从模型中获取的张量的名称.在 TensorFlow Android 演示中，这将对应于传递给 TensorFlowClassifier.initializeTensorFlow().

The main change would be to update the names of the tensor(s) that are fed into the model, and the names of the tensor(s) that are fetched from the model. In the TensorFlow Android demo, this would correspond to the inputName and outputName strings that are passed to TensorFlowClassifier.initializeTensorFlow().

GraphDef 是程序结构，通常不会在训练过程中改变.检查点是训练过程状态的快照，通常在训练过程的每一步都会发生变化.因此，TensorFlow 对这些类型的数据使用不同的存储格式，低级 API 提供了不同的方式来保存和加载它们.更高级别的库，例如 MetaGraphDef 库，Keras 和 skflow 以这些机制为基础，提供更便捷的方法来保存和恢复整个模型.>

The GraphDef is the program structure, which typically does not change through the training process. The checkpoint is a snapshot of the state of a training process, which typically changes at every step of the training process. As a result, TensorFlow uses different storage formats for these types of data, and the low-level API provides different ways to save and load them. Higher-level libraries, such as the MetaGraphDef libraries, Keras, and skflow build on these mechanisms to provide more convenient ways to save and restore an entire model.

这篇关于TensorFlow 将图形保存到文件中/从文件加载图形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！