呼叫預測(Predict) API 取得 Model 的預測結果

語法

POST http://host:port/v1/models/${MODEL_NAME}[/versions/${VERSION}|/labels/${LABEL}]:predict

API 呼叫範例

curl -d '{"instances": [1.0,2.0,5.0]}' -X POST http://localhost:8501/v1/models/saved_model_half_plus_three:predict

Request

Request 參數語法格式

{
  // (Optional) Serving signature to use.
  // If unspecifed default serving signature is used.
  "signature_name": <string>,

  // Input Tensors in row ("instances") or columnar ("inputs") format.
  // A request can have either of them but NOT both.
  "instances": <value>|<(nested)list>|<list-of-objects>
  "inputs": <value>|<(nested)list>|<object>
}

Request 範例

{
  // List of 3 scalar tensors.
  "instances": [ "foo", "bar", "baz" ]
}

{
  // List of 2 tensors each of [1, 2] shape
  "instances": [ [[1, 2]], [[3, 4]] ]
}

Response

Response 語法格式

{
  "predictions": <value>|<(nested)list>|<list-of-objects>
}

{
  "outputs": <value>|<(nested)list>|<object>
}

Response 範例

{
  "predictions": [3.5, 4.0, 5.5]
}

參考資料

- Classification: 分類
- Regression：回歸
- Predict：預測

• https://github.com/tensorflow/serving/blob/master/tensorflow_serving/apis/prediction_service.proto