> ## Documentation Index > Fetch the complete documentation index at: https://docs.identifai.net/llms.txt > Use this file to discover all available pages before exploring further. # Download an XAI map > Download the **Attention Map** or the **Attention Graph** produced by the [XAI – eXplainable AI](/classification/classify-image#xai-explainable-ai) scope for a model, for an image submitted with `with_xai=true` or for a frame of a video submitted with `with_xai=true` (use the `identifier` of the frame in `results.images`). The maps produced depend on the model: the `explanations` of the per-model results list the `type` and `method` of every map available for the model, also when the AI-powered explanations were not requested. The heatmap of the Smart Full Analysis is not an XAI map: download it with the [heatmap endpoint](./get-classification-heatmap). **Note:** This endpoint also has an alias available at `/api/classification_image/xai/{identifier}`. ## OpenAPI ````yaml get /api/classification/xai/{identifier} openapi: 3.0.0 info: title: identifAI API version: 1.0.0 description: >- API for integrating identifAI's image and video classification capabilities.
**How to obtain your API Key**To obtain your API key, follow these simple steps:

  1. **Create a user profile and log in to the platform**

    If you don't have an account yet, visit our platform and create a user profile. Once you've completed registration and verified your email, log in with your credentials.

  2. **Access the user menu**

    After logging in, click on your name or the user icon in the top-right corner of the screen.

  3. **Select the API Key option**

    In the dropdown menu that appears, select the "API Key" option. From there, you can view or generate your API key.
Now you are ready to use your API key in our API!
**How to use the API**To use the API, first make a **POST** request to start the classification. Then, use the provided ***identifier*** from the **POST** request to retrieve the results.
**1. Image Classification** * To classify an image, you need to make a **POST** request to `/api/classification` with the required parameters (for details on the parameters, refer to the *"Classification"* section below). * In the response, you will receive the ***identifier*** of the classified image. * Use the provided ***identifier*** to retrieve the classification results by making a **GET** request to `/api/classification/{identifier}` (for details on how to structure this request, see the *"Classification"* section below). * The response will contain the classification results for the image in JSON format.
In the response, you will find a ***results*** array, which includes the classification results for each model used, and a ***verdicts*** array, which contains the results for each heuristic applied.
If the classification is not yet complete, continue sending the **GET** request until the result is available.
**2. Video Classification** * To classify a video, you need to make a **POST** request to `/api/classification_video` with the required parameters (for details on the parameters, refer to the *"Classification"* section below). * In the response, you will receive the ***identifier*** of the classified video. * Use the provided ***identifier*** to retrieve the classification results by making a **GET** request to `/api/classification_video/{identifier}` (for details on how to structure this request, see the *"Classification"* section below). * The response will provide the classification results for the video in JSON format. For video classification, the video is divided into *frames*.
In the response, you will find a ***results*** array that includes the classification outcomes for each model applied to each analyzed frame, and a ***verdicts*** array containing the results for each heuristic used.
If the classification is not complete, continue sending the **GET** request until the final results are returned.
**3. All Classifications**The `/api/classifications` endpoint will return all the image classifications made by a user. This will list the results of all past image classification requests.
**4. User Management** The **POST** `/api/user` endpoint allows you to create a user, while the **DELETE** `/api/user` endpoint is used to delete a user (For the structure of the requests, see more in the user section below).
servers: - url: https://backend.identifai.net security: [] tags: - name: Classification description: Classification routes - name: Infrastructure description: Infrastructure routes - name: User description: User actions - name: Partner description: Partner actions paths: /api/classification/xai/{identifier}: get: tags: - Classification summary: Download an XAI map description: >- Download the **Attention Map** or the **Attention Graph** produced by the [XAI – eXplainable AI](/classification/classify-image#xai-explainable-ai) scope for a model, for an image submitted with `with_xai=true` or for a frame of a video submitted with `with_xai=true` (use the `identifier` of the frame in `results.images`). The maps produced depend on the model: the `explanations` of the per-model results list the `type` and `method` of every map available for the model, also when the AI-powered explanations were not requested. The heatmap of the Smart Full Analysis is not an XAI map: download it with the [heatmap endpoint](./get-classification-heatmap). **Note:** This endpoint also has an alias available at `/api/classification_image/xai/{identifier}`. parameters: - name: identifier in: path required: true schema: type: string description: The identifier of the image, or of the video frame. - name: model in: query required: true schema: type: string description: >- The name of the model that produced the map, as reported in the `model` field of the per-model results (e.g. `Revelio`). - name: method in: query required: true schema: type: string enum: - legrad - transformer_attribution - attention_rollout description: The XAI method used to compute the map. - name: type in: query required: false schema: type: string enum: - attention_map - attention_graph default: attention_map description: >- The map to download: `attention_map` (overlaid on the original content) or `attention_graph` (trend and peaks of the attention of the model). responses: '200': description: The XAI map, as an image. content: image/png: schema: type: string format: binary '400': description: Missing `model` or `method`, or invalid `type`. content: application/json: examples: missing_parameters: summary: Missing parameters value: error: 'Missing required parameters: model, method' invalid_type: summary: Invalid type value: error: >- Invalid type parameter. Allowed values: attention_map, attention_graph '401': description: Invalid or missing API key. content: application/json: example: error: Unauthorized '403': description: >- The image or frame does not exist or you are not authorized to access it. content: application/json: example: error: Unauthorized '404': description: >- No map of the requested model, method and type is available for the image or frame. content: application/json: example: error: Not found '429': description: Too many requests. content: application/json: example: error: Too Many Attempts. security: - ApiKeyAuth: [] components: securitySchemes: ApiKeyAuth: type: apiKey name: X-Api-Key in: header description: Your API key. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.