# Run ML-Assisted Labeling

You can label a project using existing model that is provided by the app or your own model. Currently, this API only support the custom API, i.e. your own model. Also, this mutation is asynchronous so the response only returns the job information.

* Mutation: [autoLabelTokenBasedProject](https://docs.datasaur.ai/#mutation-autoLabelTokenBasedProject)
* Input: [AutoLabelTokenBasedProjectInput](https://docs.datasaur.ai/#definition-AutoLabelTokenBasedProjectInput)
* Response: [Job](https://docs.datasaur.ai/#definition-Job)

### **Example Request via cURL**

```javascript
curl --location --request POST 'https://app.datasaur.ai/graphql' \
--header 'Authorization: Bearer <access_token>' \
--header 'Content-Type: application/json' \
--data-raw '{"operationName":"AutoLabelTokenBasedProjectMutation","variables":{"input":{"labelerEmail":"demo@datasaur.ai","options":{"numberOfFilesPerRequest":0,"serviceProvider":"CUSTOM"},"projectId":"HvdyIL2iwZn","targetAPI":{"endpoint":"https://custom-model.api.com/predict","secretKey":"secret"}}},"query":"mutation AutoLabelTokenBasedProjectMutation($input: AutoLabelTokenBasedProjectInput!) {  result: autoLabelTokenBasedProject(input: $input) {    id  status  progress  }}"}'
```

Below is the more readable version of request body from cURL command above.

```javascript
{
    "operationName": "AutoLabelTokenBasedProject",
    "variables": {
        "input": {
            "labelerEmail": "demo@datasaur.ai",
            "options": {
                "numberOfFilesPerRequest": 0,
                "serviceProvider": "CUSTOM"
            },
            "projectId": "HvdyIL2iwZn",
            "targetAPI": {
                "endpoint": "https://custom-model.api.com/predict",
                "secretKey": "secret"
            }
        }
    },
    "query": "mutation AutoLabelTokenBasedProjectMutation($input: AutoLabelTokenBasedProjectInput!) {  result: autoLabelTokenBasedProject(input: $input) {    id  status  progress  }}"
}
```

* **operationName**: you can fill any alphanumeric string in as the operationName. `AutoLabelTokenBasedProjectMutation` is fine as a default. Refer [this page](https://graphql.org/learn/queries/#operation-name) to organize operationName properly.
* **variables**
  * **input:** see the explanation of each variables [here](https://docs.datasaur.ai/exporttextprojectinput.doc). See the additional information for [AutoLabelProjectOptionsInput](https://docs.datasaur.ai/#definition-AutoLabelProjectOptionsInput) below.
    * Again, currently we only support CUSTOM `serviceProvider`
    * If the `numberOfFilesPerRequest` is 0, then it will request all lines in a document. If it's more than 0, the backend will batch the request based on how many lines for each request.
* **query:** Copy it from cURL example.

### Example Response

Here is the response when you execute the cURL command above.

```javascript
{
    "data": {
        "result": {
            "id": "AutoLabelQueue:dcbf6ed3-3841-46c1-8f6a-2affb53ac637",
            "status": "IN_PROGRESS",
            "progress": 50
        }
    },
    "extensions": {}
}
```

This doesn't mean the process is finished. To make sure the job is finished or to get any other information, please see [this page](https://docs.datasaur.ai/api/get-data/check-job).


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.datasaur.ai/api/run-ml-assisted-labeling.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.