使用现有 frozen_interface_graph.pb 和 label_map.pbtxt 部署 TFX

Question

我用 fasterR-CNN 网络训练了一个目标检测模型，训练后有 frozen_interface_graph.pb 和 label_map.pbtxt。我想将它部署为 RESTAPI 服务器，以便可以从没有 Tensorflow 的系统中调用它。那是我遇到 TFX 的时候。

如何使用 TFX tensorflow-model-server 加载此模型并托管 RESTAPI，以便我可以发送图像作为 POST 请求进行预测？

https://www.tensorflow.org/tfx/tutorials/serving/rest_simple This is what I found as a reference, but the models are of a different format than what I have currently. Is there any mechanism in which I can reuse the model I currently have or will I have to retrain using Keras and deploy as shown in the reference.

Answer 1

要为 TFX 重用您的模型，冻结图 需要指定服务签名。尝试使用下面的代码将您的模型转换为 savedmodel 格式，成功创建了带有标签集 "serve".

的 savedmodel.pb 文件

import tensorflow as tf
from tensorflow.python.saved_model import signature_constants
from tensorflow.python.saved_model import tag_constants

export_dir = './saved'
graph_pb = 'frozen_inference_graph.pb'

builder = tf.saved_model.builder.SavedModelBuilder(export_dir)

with tf.gfile.GFile(graph_pb, "rb") as f:
    graph_def = tf.GraphDef()
    graph_def.ParseFromString(f.read())

sigs = {}

with tf.Session(graph=tf.Graph()) as sess:
    # name="" is important to ensure we don't get spurious prefixing
    tf.import_graph_def(graph_def, name="")
    g = tf.get_default_graph()
    sess.graph.get_operations()
    inp = g.get_tensor_by_name("image_tensor:0")
    outputs = {}
    outputs["detection_boxes"] = g.get_tensor_by_name('detection_boxes:0')
    outputs["detection_scores"] = g.get_tensor_by_name('detection_scores:0')
    outputs["detection_classes"] = g.get_tensor_by_name('detection_classes:0')
    outputs["num_detections"] = g.get_tensor_by_name('num_detections:0')

    output_tensor = tf.concat([tf.expand_dims(t, 0) for t in outputs], 0)


    sigs[signature_constants.DEFAULT_SERVING_SIGNATURE_DEF_KEY] = \
        tf.saved_model.signature_def_utils.predict_signature_def(
            {"in": inp}, {"out": out})

    sigs["predict_images"] = \
    tf.saved_model.signature_def_utils.predict_signature_def(
        {"in": inp}, {"out": output_tensor} )

    builder.add_meta_graph_and_variables(sess,
                                         [tag_constants.SERVING],
                                         signature_def_map=sigs)

builder.save().

我们已经通过预测您提供的样本图像测试了转换后的模型。结果没有显示任何预测，这可能意味着转换方法没有按预期工作。

关于你的问题：

"Is there any mechanism in which I can reuse the model I currently have or will I have to retrain using Keras and deploy as shown in the reference?"

有了这个结果，最好只使用 Keras 来重新训练你的模型作为你问题的答案，因为转换或重用你的冻结图模型不会是解决方案。您的模型未保存为模型提供服务所需的变量，并且模型格式不适合生产环境。是的，这是遵循 official documentation 的最佳方式，因为您可以确信这会奏效。

使用现有 frozen_interface_graph.pb 和 label_map.pbtxt 部署 TFX

deploy TFX with existing frozen_interface_graph.pb and label_map.pbtxt

python

tensorflow

tfx