Question

我使用tf.estimator .method export_savedmodel保存模型，如下所示：

export_dir="exportModel/"

feature_spec = tf.feature_column.make_parse_example_spec(feature_columns)

input_receiver_fn = tf.estimator.export.build_parsing_serving_input_receiver_fn(feature_spec)

classifier.export_savedmodel(export_dir, input_receiver_fn, as_text=False, checkpoint_path="Model/model.ckpt-400")

如何导入此保存的模型并用于预测？

Answer 1

我试图搜索一个很好的基础示例，但看起来文档和示例对于此主题来说有点分散。因此，让我们从一个基本示例开始：tf.estimator quickstart。

该特定示例实际上并未导出模型，因此我们这样做（不需要用例1）：

def serving_input_receiver_fn():
  """Build the serving inputs."""
  # The outer dimension (None) allows us to batch up inputs for
  # efficiency. However, it also means that if we want a prediction
  # for a single instance, we'll need to wrap it in an outer list.
  inputs = {"x": tf.placeholder(shape=[None, 4], dtype=tf.float32)}
  return tf.estimator.export.ServingInputReceiver(inputs, inputs)

export_dir = classifier.export_savedmodel(
    export_dir_base="/path/to/model",
    serving_input_receiver_fn=serving_input_receiver_fn)

此代码上的巨大星号：TensorFlow 1.3中似乎存在一个错误，它不允许您在＆＃34; canned＆＃34;上进行上述导出。估计器（例如DNNClassifier）。有关解决方法，请参阅＆＃34;附录：解决方法＆＃34;部分。

下面的代码引用export_dir（导出步骤的返回值）以强调它是不＆＃34; / path / to / model＆＃34;，而是，该目录的子目录，其名称为时间戳。

用例1：在与培训相同的过程中执行预测

这是一种sci-kit学习类型的经验，并已通过示例进行了示例。为了完整性＆＃39;为此，你只需在训练有素的模型上调用predict：

classifier.train(input_fn=train_input_fn, steps=2000)
# [...snip...]
predictions = list(classifier.predict(input_fn=predict_input_fn))
predicted_classes = [p["classes"] for p in predictions]

用例2：将SavedModel加载到Python / Java / C ++中并执行预测

Python客户端

如果你想在Python中进行预测，也许最容易使用的是SavedModelPredictor。在将使用SavedModel的Python程序中，我们需要这样的代码：

from tensorflow.contrib import predictor

predict_fn = predictor.from_saved_model(export_dir)
predictions = predict_fn(
    {"x": [[6.4, 3.2, 4.5, 1.5],
           [5.8, 3.1, 5.0, 1.7]]})
print(predictions['scores'])

Java客户端

package dummy;

import java.nio.FloatBuffer;
import java.util.Arrays;
import java.util.List;

import org.tensorflow.SavedModelBundle;
import org.tensorflow.Session;
import org.tensorflow.Tensor;

public class Client {

  public static void main(String[] args) {
    Session session = SavedModelBundle.load(args[0], "serve").session();

    Tensor x =
        Tensor.create(
            new long[] {2, 4},
            FloatBuffer.wrap(
                new float[] {
                  6.4f, 3.2f, 4.5f, 1.5f,
                  5.8f, 3.1f, 5.0f, 1.7f
                }));

    // Doesn't look like Java has a good way to convert the
    // input/output name ("x", "scores") to their underlying tensor,
    // so we hard code them ("Placeholder:0", ...).
    // You can inspect them on the command-line with saved_model_cli:
    //
    // $ saved_model_cli show --dir $EXPORT_DIR --tag_set serve --signature_def serving_default
    final String xName = "Placeholder:0";
    final String scoresName = "dnn/head/predictions/probabilities:0";

    List<Tensor> outputs = session.runner()
        .feed(xName, x)
        .fetch(scoresName)
        .run();

    // Outer dimension is batch size; inner dimension is number of classes
    float[][] scores = new float[2][3];
    outputs.get(0).copyTo(scores);
    System.out.println(Arrays.deepToString(scores));
  }
}

C ++客户端

您可能希望将tensorflow::LoadSavedModel与Session一起使用。

#include <unordered_set>
#include <utility>
#include <vector>

#include "tensorflow/cc/saved_model/loader.h"
#include "tensorflow/core/framework/tensor.h"
#include "tensorflow/core/public/session.h"

namespace tf = tensorflow;

int main(int argc, char** argv) {
  const string export_dir = argv[1];

  tf::SavedModelBundle bundle;
  tf::Status load_status = tf::LoadSavedModel(
      tf::SessionOptions(), tf::RunOptions(), export_dir, {"serve"}, &bundle);
  if (!load_status.ok()) {
    std::cout << "Error loading model: " << load_status << std::endl;
    return -1;
  }

  // We should get the signature out of MetaGraphDef, but that's a bit
  // involved. We'll take a shortcut like we did in the Java example.
  const string x_name = "Placeholder:0";
  const string scores_name = "dnn/head/predictions/probabilities:0";

  auto x = tf::Tensor(tf::DT_FLOAT, tf::TensorShape({2, 4}));
  auto matrix = x.matrix<float>();
  matrix(0, 0) = 6.4;
  matrix(0, 1) = 3.2;
  matrix(0, 2) = 4.5;
  matrix(0, 3) = 1.5;
  matrix(0, 1) = 5.8;
  matrix(0, 2) = 3.1;
  matrix(0, 3) = 5.0;
  matrix(0, 4) = 1.7;

  std::vector<std::pair<string, tf::Tensor>> inputs = {{x_name, x}};
  std::vector<tf::Tensor> outputs;

  tf::Status run_status =
      bundle.session->Run(inputs, {scores_name}, {}, &outputs);
  if (!run_status.ok()) {
    cout << "Error running session: " << run_status << std::endl;
    return -1;
  }

  for (const auto& tensor : outputs) {
    std::cout << tensor.matrix<float>() << std::endl;
  }
}

使用案例3：使用TensorFlow服务模拟服务

以适合Classification model的方式导出模型要求输入为tf.Example个对象。以下是我们如何为TensorFlow服务导出模型：

def serving_input_receiver_fn():
  """Build the serving inputs."""
  # The outer dimension (None) allows us to batch up inputs for
  # efficiency. However, it also means that if we want a prediction
  # for a single instance, we'll need to wrap it in an outer list.
  example_bytestring = tf.placeholder(
      shape=[None],
      dtype=tf.string,
  )
  features = tf.parse_example(
      example_bytestring,
      tf.feature_column.make_parse_example_spec(feature_columns)
  )
  return tf.estimator.export.ServingInputReceiver(
      features, {'examples': example_bytestring})

export_dir = classifier.export_savedmodel(
    export_dir_base="/path/to/model",
    serving_input_receiver_fn=serving_input_receiver_fn)

有关如何设置TensorFlow服务的更多说明，请参阅TensorFlow服务文档，因此我只在此处提供客户端代码：

  # Omitting a bunch of connection/initialization code...
  # But at some point we end up with a stub whose lifecycle
  # is generally longer than that of a single request.
  stub = create_stub(...)

  # The actual values for prediction. We have two examples in this
  # case, each consisting of a single, multi-dimensional feature `x`.
  # This data here is the equivalent of the map passed to the 
  # `predict_fn` in use case #2.
  examples = [
    tf.train.Example(
      features=tf.train.Features(
        feature={"x": tf.train.Feature(
          float_list=tf.train.FloatList(value=[6.4, 3.2, 4.5, 1.5]))})),
    tf.train.Example(
      features=tf.train.Features(
        feature={"x": tf.train.Feature(
          float_list=tf.train.FloatList(value=[5.8, 3.1, 5.0, 1.7]))})),
  ]

  # Build the RPC request.
  predict_request = predict_pb2.PredictRequest()
  predict_request.model_spec.name = "default"
  predict_request.inputs["examples"].CopyFrom(
      tensor_util.make_tensor_proto(examples, tf.float32))

  # Perform the actual prediction.
  stub.Predict(request, PREDICT_DEADLINE_SECS)

请注意，examples中引用的密钥predict_request.inputs需要与导出时serving_input_receiver_fn中使用的密钥匹配（参见构造函数ServingInputReceiver在那段代码中。）

附录：解决TF 1.3中的预制模型的出口

TensorFlow 1.3中似乎存在一个错误，其中固定模型无法正确导出用例2（＆＃34;自定义＆＃34;估算器不存在该问题）。这是一个解决方法，它包装了一个DNNClassifier以使工作正常，特别是对于Iris示例：

# Build 3 layer DNN with 10, 20, 10 units respectively.
class Wrapper(tf.estimator.Estimator):
  def __init__(self, **kwargs):
    dnn = tf.estimator.DNNClassifier(**kwargs)

    def model_fn(mode, features, labels):
      spec = dnn._call_model_fn(features, labels, mode)
      export_outputs = None
      if spec.export_outputs:
        export_outputs = {
           "serving_default": tf.estimator.export.PredictOutput(
                  {"scores": spec.export_outputs["serving_default"].scores,
                   "classes": spec.export_outputs["serving_default"].classes})}

      # Replace the 3rd argument (export_outputs)
      copy = list(spec)
      copy[4] = export_outputs
      return tf.estimator.EstimatorSpec(mode, *copy)

    super(Wrapper, self).__init__(model_fn, kwargs["model_dir"], dnn.config)

classifier = Wrapper(feature_columns=feature_columns,
                     hidden_units=[10, 20, 10],
                     n_classes=3,
                     model_dir="/tmp/iris_model")

Answer 2

我不认为罐装Estimators存在错误（或者更确切地说，如果有的话，它已被修复）。我能够使用Python成功导出罐装估算器模型并将其导入Java。

以下是导出模型的代码：

   QProcess *maxwell_process = new QProcess;

    maxwell_process->start(maxwellExePath, args);

    connect(maxwell_process, static_cast<void (QProcess::*)(int, QProcess::ExitStatus)>(&QProcess::finished),
        [&](int exitCode, QProcess::ExitStatus exitStatus){
        printf("process stopped");
        this->stopped = true;
    });
    t_ini = clock();
    while (1) {

        Sleep(100);
        t_fin = clock();
        secs = (double)(t_fin - t_ini);
        secs = secs / CLOCKS_PER_SEC;
        if (solution->getTimeOutValue() == -1) {
            std::cout<<"init population time"<<std::endl;
            if (this->stopped) {
                disconnect(maxwell_process);
                std::cout<<"init population time "<<secs<<std::endl;
                solution->setTime(secs);
                break;
            }
        } else {
            if (secs > solution->getTimeOutValue()) {
                disconnect(maxwell_process);
                printf("timeout");
                maxwell_process->kill();
                solution->setTime(secs);
                break;
            }
        }
                QCoreApplication::processEvents();
    }

要在Java中导入模型，我使用上面的rhaertel80提供的Java客户端代码，它可以工作。希望这也能回答Ben Fowler的上述问题。

Answer 3

TensorFlow团队似乎不同意版本1.3中存在使用预设估算器导出模型的错误＃2。我在这里提交了一个错误报告： https://github.com/tensorflow/tensorflow/issues/13477

我从TensorFlow收到的响应是输入必须只是单个字符串张量。似乎可能有一种方法可以使用序列化的TF.examples将多个功能合并到单个字符串张量中，但我还没有找到一个明确的方法来执行此操作。如果有人有代码显示如何做到这一点，我会很感激。

Answer 4

您需要使用tf.contrib.export_savedmodel导出已保存的模型，并且需要定义输入接收器函数以将输入传递给。稍后您可以从磁盘加载已保存的模型（通常为saved.model.pb）并提供服务。

TensorFlow: How to predict from a SavedModel?

如何使用tf.estimator导入已保存的Tensorflow模型序列并预测输入数据

4 个答案: