张量流服务错误：参数无效：JSON 对象：没有命名输入

我正在尝试使用 Amazon Sagemaker 训练模型，并且希望使用 Tensorflow 服务来为其提供服务。为了实现这一目标，我将模型下载到 Tensorflow 服务 docker，并尝试从那里提供服务。

Sagemaker 的训练和评估阶段已顺利完成，但当我将模型加载到 Tensorflow 服务服务器并尝试调用它时，我收到 Tensorflow 服务错误，表明我的模型没有定义的输入。可以看到模型正在服务的Tensorflow服务服务器。

出于调试目的，我尝试使用 Sagemaker 提供服务，但得到的只是一条模糊的错误消息，表明我调用端点时出错。

我认为问题在于我没有很好地定义serving_input_fn 或调用它错误或两者兼而有之。有人可以帮忙吗？

Tensorflow 服务服务器调用curl：

curl -d '{"instances": [{"col3": 1.0}]}' -X POST http://localhost:8501/v1/models/test_model:predict

我从 Tensorflow 服务收到的错误：

{ "error": "Failed to process element: 0 key: col3 of \'instances\' list. Error: Invalid argument: JSON object: does not have named input: col3" }%    

Sagemaker的训练python文件：

import os
import tensorflow as tf
from tensorflow.python.ops import nn


TRAIN_FILENAME = 'test.csv'
TEST_FILENAME = 'train.csv'

NODES_IN_LAYER = 6
LAYERS_NUM = 10
NUM_LINES_TO_SKIP = 1

CSV_COLUMNS = ['col1', 'col2', 'col3', 'col4', 'col5', 'col6', 'col7', 'col8', 'label']
RECORDS_DEFAULTS = [[0], [0], [0.0], [0.0], [0], [0.0], [0.0], [0], [0.0]]

BATCH_SIZE = 32

FEATURE_SPEC = {
    'col3': tf.FixedLenFeature(dtype=tf.float32, shape=[]),
}


def estimator_fn(run_config, params):
    feature_columns = [
        tf.feature_column.numeric_column('col3')]
    return tf.estimator.DNNRegressor(feature_columns=feature_columns,
                                     hidden_units=[NODES_IN_LAYER] * LAYERS_NUM,
                                     activation_fn=nn.tanh,
                                     config=run_config)


def serving_input_fn(params):
    return tf.estimator.export.build_raw_serving_input_receiver_fn(FEATURE_SPEC)


def train_input_fn(training_dir, params):
    """Returns input function that would feed the model during training"""
    return _generate_input_fn(training_dir, TRAIN_FILENAME)


def eval_input_fn(training_dir, params):
    """Returns input function that would feed the model during evaluation"""
    return _generate_input_fn(training_dir, TEST_FILENAME)


def parse_csv(line):
    columns = tf.decode_csv(line, record_defaults=RECORDS_DEFAULTS)
    line_features = dict(zip(CSV_COLUMNS, columns))
    line_label = line_features.pop('label')
    return {'col3': line_features.pop('col3')}, line_label


def _generate_input_fn(training_dir, training_filename):
    filename = os.path.join(training_dir, training_filename)
    dataset = tf.data.TextLineDataset(filename)
    dataset = dataset.skip(NUM_LINES_TO_SKIP).map(parse_csv).batch(BATCH_SIZE)
    return dataset

当模型的输入与您提供的输入不匹配时，就会出现此错误。

最好的方法是通过发出如下 get 请求来检查服务模型的输入：

http://<ip>:8501/v1/models/bilstm/metadata

它会返回类似的输出

{
    "model_spec": {
        "name": "bilstm",
        "signature_name": "",
        "version": "1"
    },
    "metadata": {
        "signature_def": {
            "signature_def": {
                "serving_default": {
                    "inputs": {
                        "sequence_length": {
                            "dtype": "DT_INT32",
                            "tensor_shape": {
                                "dim": [
                                    {
                                        "size": "-1",
                                        "name": ""
                                    }
                                ],
                                "unknown_rank": false
                            },
                            "name": "sequence_lengths:0"
                        },
                        "word_ids": {
                            "dtype": "DT_INT32",
                            "tensor_shape": {
                                "dim": [
                                    {
                                        "size": "-1",
                                        "name": ""
                                    },
                                    {
                                        "size": "-1",
                                        "name": ""
                                    }
                                ],
                                "unknown_rank": false
                            },
                            "name": "word_ids:0"
                        },
                        "lr": {
                            "dtype": "DT_FLOAT",
                            "tensor_shape": {
                                "dim": [],
                                "unknown_rank": false
                            },
                            "name": "lr:0"
                        },
                        "word_lengths": {
                            "dtype": "DT_INT32",
                            "tensor_shape": {
                                "dim": [
                                    {
                                        "size": "-1",
                                        "name": ""
                                    },
                                    {
                                        "size": "-1",
                                        "name": ""
                                    }
                                ],
                                "unknown_rank": false
                            },
                            "name": "word_lengths:0"
                        },
                        "char_ids": {
                            "dtype": "DT_INT32",
                            "tensor_shape": {
                                "dim": [
                                    {
                                        "size": "-1",
                                        "name": ""
                                    },
                                    {
                                        "size": "-1",
                                        "name": ""
                                    },
                                    {
                                        "size": "-1",
                                        "name": ""
                                    }
                                ],
                                "unknown_rank": false
                            },
                            "name": "char_ids:0"
                        },
                        "dropout": {
                            "dtype": "DT_FLOAT",
                            "tensor_shape": {
                                "dim": [],
                                "unknown_rank": false
                            },
                            "name": "dropout:0"
                        }
                    },
                    "outputs": {
                        "scores": {
                            "dtype": "DT_FLOAT",
                            "tensor_shape": {
                                "dim": [
                                    {
                                        "size": "-1",
                                        "name": ""
                                    },
                                    {
                                        "size": "-1",
                                        "name": ""
                                    },
                                    {
                                        "size": "30",
                                        "name": ""
                                    }
                                ],
                                "unknown_rank": false
                            },
                            "name": "bi-lstm-crf/output_ff/BiasAdd:0"
                        }
                    },
                    "method_name": "tensorflow/serving/predict"
                }
            }
        }
    }
}