Sentence Vector

Introduction

This API is used to return the corresponding sentence vector when you enter a sentence.

For details about endpoints, see Endpoints.

Calling NLP APIs will incur fees. NLP packages are classified into the basic and domain-specific editions. When purchasing a package, view the APIs supported by the basic package and domain-specific packages in the Natural Language Processing Price Calculator.

URI

  • URI format
    POST /v1/{project_id}/nlp-fundamental/sentence-embedding
  • Parameter description
    Table 1 URI parameters

    Parameter

    Mandatory

    Description

    project_id

    Yes

    Project ID. For details about how to obtain the project ID, see Obtaining a Project ID.

Request

Table 2 describes the request parameters.

Table 2 Request parameters

Parameter

Type

Mandatory

Description

sentences

Array of strings

Yes

Text list. The text is encoded using UTF-8 and contains 1 to 512 characters. The text list contains 1 to 1,000 text data records. Chinese is supported currently.

domain

String

No

Name of the model used for computing the sentence vector. Currently, only the default value general is available.

Response

Table 3 describes the response parameters.

Table 3 Response parameters

Parameter

Type

Description

vectors

Array of floats

Sentence vector result list. Sentence vectors are returned according to the input sentence sequence. The default dimension of the sentence vector is 100.

error_code

String

Error code when the API fails to be called. For details, see Error Code.

The parameter is not included when the API call succeeds.

error_msg

String

Error message returned when the API fails to be called.

The parameter is not included when the API call succeeds.

Example

  • Example request
    POST https://{endpoint}/v1/{project_id}/nlp-fundamental/sentence-embedding
    
    Request Header:
        Content-Type: application/json
        X-Auth-Token: MIINRwYJKoZIhvcNAQcCoIINODCCDTQCAQExDTALBglghkgBZQMEAgEwgguVBgkqhkiG...
    
    Request Body:
        {
    ...."sentences":["Zhang San is in Beijing today.","Li Si is in Beijing today."],
            "domain":"general"
        }
  • Example response
    • Successful response example
      {
        "vectors": [
          [0.1331, 0.0488, 0.2441, 0.2514, -0.6771, 0.4782, 0.6759, 0.015, 0.0064, -0.6326, 0.3958, -0.6848, 0.1118, -0.1391, 0.4804, 0.9294, 0.1004, 0.2414, 0.2477, -0.8162, 1.2052, -0.6719, -0.47, -0.1946, -0.0606, 0.473, 0.0247, -0.3857, 1.1637, -0.6092, -0.5512, -0.2389, -0.2168, 0.1673, -0.4124, -0.1196, -0.7147, 1.1774, -0.8166, 0.1285, -0.3136, 0.4687, -0.5939, -0.4579, 0.1857, 0.049, -0.5936, -0.4554, -0.1878, 0.017],
          [0.0833, -0.0731, 0.298, 0.0085, -0.6858, 0.529, 0.887, 0.1772, -0.118, -0.7559, 0.1995, -0.6415, 0.3014, 0.2061, 0.9727, 0.9089, 0.1603, 0.3773, -0.146, -0.6429, 1.4808, -0.7797, -0.6061, -0.0854, -0.1324, 0.3183, 0.3378, -0.4552, 1.4929, -0.7543, -0.6089, -0.1906, -0.1892, 0.0628, -0.4675, -0.2478, -0.7632, 1.1876, -1.0734, -0.0954, -0.2896, 0.5757, -0.5601, -0.2595, 0.3831, 0.4729, -0.8736, -0.4378, -0.2519, 0.0448]
        ]
      }
    • Failed response example
      {
          "error_code": "NLP.0301",
          "error_msg": "argument valid error: sentence must not be blank and sentence length 1-512"
      }

Status Code

For details about status codes, see Status Code.

Error Code

For details about error codes, see Error Code.