将元数据添加到Tensorflow服务API调用

时间:2018-11-15 23:39:25

标签: tensorflow-serving

是否可以将元数据添加到服务servable的张量流中,以便该元数据也填充在可服务对象的响应中?

如果我对文件结构有把握:

my_servable/ 
           1541778457/ 
                     variables/ 
                     saved_model.pb 

例如:

```
outputs {
  key: "classes"
  value {
    dtype: DT_STRING
    tensor_shape {
      dim {
        size: 8
      }
    }
    string_val: "a"
    string_val: "b"
    string_val: "c"
    string_val: "d"
    string_val: "e"
    string_val: "f"
    string_val: "g"
    string_val: "h"
  }
}
outputs {
  key: "scores"
  value {
    dtype: DT_FLOAT
    tensor_shape {
      dim {
        size: 1
      }
      dim {
        size: 8
      }
    }
    float_val: 1.212528104588273e-06
    float_val: 5.094948463124638e-08
    float_val: 0.0009737954242154956
    float_val: 0.9988483190536499
    float_val: 3.245145592245535e-07
    float_val: 0.00010837535955943167
    float_val: 4.101086960872635e-05
    float_val: 2.676981057447847e-05
  }
}
model_spec {
  name: "my_model"
  version {
    value: 1541778457
  }
  signature_name: "prediction"
}

如果对于生成此可服务项的代码(例如f6ca434910504532a0d50dfd12f22d4c)来说,我有类似git哈希或唯一标识符的代码,是否可以在客户端请求中获取此数据?

理想情况是:

```
outputs {
  key: "classes"
  value {
    dtype: DT_STRING
    tensor_shape {
      dim {
        size: 8
      }
    }
    string_val: "a"
    string_val: "b"
    string_val: "c"
    string_val: "d"
    string_val: "e"
    string_val: "f"
    string_val: "g"
    string_val: "h"
  }
}
outputs {
  key: "scores"
  value {
    dtype: DT_FLOAT
    tensor_shape {
      dim {
        size: 1
      }
      dim {
        size: 8
      }
    }
    float_val: 1.212528104588273e-06
    float_val: 5.094948463124638e-08
    float_val: 0.0009737954242154956
    float_val: 0.9988483190536499
    float_val: 3.245145592245535e-07
    float_val: 0.00010837535955943167
    float_val: 4.101086960872635e-05
    float_val: 2.676981057447847e-05
  }
}
model_spec {
  name: "my_model"
  version {
    value: 1541778457
  }
  hash {
    value: f6ca434910504532a0d50dfd12f22d4c
 }
  signature_name: "prediction"
}

我尝试将目录从1541778457更改为哈希,但这给出了:

W tensorflow_serving/sources/storage_path/file_system_storage_path_source.cc:268] No versions of servable default found under base path

1 个答案:

答案 0 :(得分:0)

我想您可以通过两种方式解决此问题。如果您想更改文件夹名称以使其正常工作,请记住在这种情况下,该文件夹名称描述您的模型版本,我认为该版本必须为整数。因此,我假设您需要将哈希转换为二进制或十进制,然后在收到时将其转换回。

我认为,更好的解决方案是您是否能够更改模型并添加包含哈希的变量。并将其添加到模型signature_def。在python中看起来像:

// create your field
hash = tf.placeholder("f6ca434910504532a0d50dfd12f22d4c",tf.string, name="HASH")

// build tensor
hash_info = tf.saved_model.utils.build_tensor_info(hash)

// add hash_info in your output in signature_def

// then you should be able to receive that data in your request