huggingface-api

A simple wrapper for the Huggingface api.

Example

Javascript

const hf = require('huggingface-api')

hf.request({
    text: 'My name is Jeff and',
    model: 'EleutherAI/gpt-neo-2.7B',
    api_key: 'api_[API KEY REDACTED]',
    return_type: 'STRING'
}).then((data) => {
    console.log(data)
})

Typescript

import * as hf from 'huggingface-api'

hf.request({
    text: 'My name is Jeff and',
    model: 'EleutherAI/gpt-neo-2.7B',
    api_key: 'api_[API KEY REDACTED]',
    return_type: 'STRING'
}).then((data) => {
    console.log(data)
})

In-depth

request

export declare function request(data: Data): Promise<any>

Data is an argument containing the request data:

export interface Data {
    text?: string
    file?: string
    api_key: string
    model: string
    return_type: 'FULL' | 'OBJ' | 'STRING'
}

text: The text you want to submit. file: If you want to read the text from a file, use this.

NOTE: If both text and file are used, text has a priority

api_key: Your huggingface api key. This can be found in here model: Name of the model.

If you are on the page of the model, the name eg. the EleutherAI/gpt-neo-2.7B part of huggingface.co/EleutherAI/gpt-neo-2.7B

return_type: How is the data returned: FULL

[
    {
        generated_text: 'Something...'
    }
]

OBJ

{
    generated_text: 'Something...'
}

STRING

'Something...'