Calculating the Number of Tokens
Function
Calculates the number of tokens. By default, the input text is segmented.
Format
Operation type: POST
URL: https://{ip}:{port}/v1/tokenizer
Replace {ip} and {port} with the IP address and port number of the service plane, that is, ipAddress and port.
Request Parameters
Parameter |
Mandatory/Optional |
Description |
Value |
|---|---|---|---|
inputs |
Mandatory |
Input character string whose number of tokens needs to be calculated. |
The value is of the string type. The value range is [0, 4 × 1024 × 1024]. |
do_decode |
Optional |
Determines the content returned in the tokens field. The default value is true. |
Bool type.
|
Usage Example
- Example 1:
Request example:
POST https://{ip}:{port}/v1/tokenizerRequest body:
{ "inputs": "xxxxxx" }Response example:{ "token_number": 1234, "tokens": ["abc", "xyz", ...] }
- Example 2:
Request example:
POST https://{ip}:{port}/v1/tokenizerRequest body:
{ "inputs": "xxxxxx", "do_decode": false }Response example:{ "token_number": 1234, "tokens": [] }
Output Description
Return Value |
Type |
Description |
|---|---|---|
token_number |
Integer |
Number of tokens. |
tokens |
String array |
List of parsed tokens. If a Chinese character consists of multiple tokens, the tokens are combined into one Chinese character. Therefore, the number of this array may be smaller than the value of token_number. |