# Machine learning components
<a name="machine-learning-components"></a>

AWS IoT Greengrass provides the following machine learning components that you can deploy to supported devices to [perform machine learning inference](perform-machine-learning-inference.md) using models trained in Amazon SageMaker AI or with your own pre-trained models that are stored in Amazon S3. <a name="ml-component-types"></a>

AWS provides the following categories of machine learning components:
+ **Model component**—Contains machine learning models as Greengrass artifacts.
+ **Runtime component**—Contains the script that installs the machine learning framework and its dependencies on the Greengrass core device.
+ **Inference component**—Contains the inference code and includes component dependencies to install the machine learning framework and download pre-trained machine learning models.

You can use the sample inference code and pre-trained models in the AWS-provided machine learning components to perform image classification and object detection using DLR and TensorFlow Lite. To perform custom machine learning inference with your own models that are stored in Amazon S3, or to use a different machine learning framework, you can use the recipes of these public components as templates to create custom machine learning components. For more information, see [Customize your machine learning components](ml-customization.md).

AWS IoT Greengrass also includes an AWS-provided component to manage the installation and lifecycle of the SageMaker AI Edge Manager agent on Greengrass core devices. With SageMaker AI Edge Manager, you can use Amazon SageMaker AI Neo-compiled models directly on your core device. For more information, see [Use Amazon SageMaker AI Edge Manager on Greengrass core devices](use-sagemaker-edge-manager.md).

The following table lists the machine learning components that are available in AWS IoT Greengrass. 

**Note**  <a name="component-nucleus-dependency-update-note"></a>
Several AWS-provided components depend on specific minor versions of the Greengrass nucleus. Because of this dependency, you need to update these components when you update the Greengrass nucleus to a new minor version. For information about the specific versions of the nucleus that each component depends on, see the corresponding component topic. For more information about updating the nucleus, see [Update the AWS IoT Greengrass Core software (OTA)](update-greengrass-core-v2.md).

<a name="component-table-type-description"></a>When a component has a component type of both generic and Lambda, the current version of the component is the generic type and a previous version of the component is the Lambda type.


| Component | Description | [Component type](develop-greengrass-components.md#component-types) | Supported OS | [Open source](open-source.md) | 
| --- | --- | --- | --- | --- | 
| [SageMaker AI Edge Manager](sagemaker-edge-manager-component.md) | Deploys the Amazon SageMaker AI Edge Manager agent on the Greengrass core device. | Generic | Linux, Windows | No | 
| [DLR image classification](dlr-image-classification-component.md) | Inference component that uses the DLR image classification model store and the DLR runtime component as dependencies to install DLR, download sample image classification models, and perform image classification inference on supported devices. | Generic | Linux, Windows | No | 
| [DLR object detection](dlr-object-detection-component.md) | Inference component that uses the DLR object detection model store and the DLR runtime component as dependencies to install DLR, download sample object detection models, and perform object detection inference on supported devices. | Generic | Linux, Windows | No | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | Model component that contains sample ResNet-50 image classification models as Greengrass artifacts. | Generic | Linux, Windows | No | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | Model component that contains sample YOLOv3 object detection models as Greengrass artifacts. | Generic | Linux, Windows | No | 
| [DLR runtime](dlr-component.md) | Runtime component that contains an installation script that is used to install DLR and its dependencies on the Greengrass core device. | Generic | Linux, Windows | No | 
| [TensorFlow Lite image classification](tensorflow-lite-image-classification-component.md) | Inference component that uses the TensorFlow Lite image classification model store and the TensorFlow Lite runtime component as dependencies to install TensorFlow Lite, download sample image classification models, and perform image classification inference on supported devices. | Generic | Linux, Windows | No | 
| [TensorFlow Lite object detection](tensorflow-lite-object-detection-component.md) | Inference component that uses the TensorFlow Lite object detection model store and the TensorFlow Lite runtime component as dependencies to install TensorFlow Lite, download sample object detection models, and perform object detection inference on supported devices. | Generic | Linux, Windows | No | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | Model component that contains a sample MobileNet v1 model as a Greengrass artifact. | Generic | Linux, Windows | No | 
| [TensorFlow Lite object detection model store](tensorflow-lite-object-detection-model-store-component.md) | Model component that contains a sample Single Shot Detection (SSD) MobileNet model as a Greengrass artifact. | Generic | Linux, Windows | No | 
| [TensorFlow Lite runtime](tensorflow-lite-component.md) | Runtime component that contains an installation script that is used to install TensorFlow Lite and its dependencies on the Greengrass core device. | Generic | Linux, Windows | No | 

# SageMaker AI Edge Manager
<a name="sagemaker-edge-manager-component"></a>

**Important**  
SageMaker AI Edge Manager was discontinued on April 26th, 2024. For more information about continuing to deploy your models to edge devices, see [SageMaker AI Edge Manager end of life](https://docs.aws.amazon.com/sagemaker/latest/dg/edge-eol.html).

The Amazon SageMaker AI Edge Manager component (`aws.greengrass.SageMakerEdgeManager`) installs the SageMaker AI Edge Manager agent binary. 

SageMaker AI Edge Manager provides model management for edge devices so you can optimize, secure, monitor, and maintain machine learning models on fleets of edge devices. The SageMaker AI Edge Manager component installs and manages the lifecycle of the SageMaker AI Edge Manager agent on your core device. You can also use SageMaker AI Edge Manager to package and use SageMaker AI Neo-compiled models as model components on Greengrass core devices. For more information about using SageMaker AI Edge Manager agent on your core device, see [Use Amazon SageMaker AI Edge Manager on Greengrass core devices](use-sagemaker-edge-manager.md).

SageMaker AI Edge Manager component v1.3.x installs Edge Manager agent binary v1.20220822.836f3023. For more information about Edge Manager agent binary versions, see [Edge Manager Agent](https://docs.aws.amazon.com/sagemaker/latest/dg/edge-device-fleet-about).

**Note**  
The SageMaker AI Edge Manager component is available only in the following AWS Regions:  
US East (Ohio)
US East (N. Virginia)
US West (Oregon)
EU (Frankfurt)
EU (Ireland)
Asia Pacific (Tokyo)

**Topics**
+ [Versions](#sagemaker-edge-manager-component-versions)
+ [Type](#sagemaker-edge-manager-component-type)
+ [Operating system](#sagemaker-edge-manager-component-os-support)
+ [Requirements](#sagemaker-edge-manager-component-requirements)
+ [Dependencies](#sagemaker-edge-manager-component-dependencies)
+ [Configuration](#sagemaker-edge-manager-component-configuration)
+ [Local log file](#sagemaker-edge-manager-component-log-file)
+ [Changelog](#sagemaker-edge-manager-component-changelog)

## Versions
<a name="sagemaker-edge-manager-component-versions"></a>

This component has the following versions:
+ 1.3.x
+ 1.2.x
+ 1.1.x
+ 1.0.x

## Type
<a name="sagemaker-edge-manager-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="sagemaker-edge-manager-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="sagemaker-edge-manager-component-requirements"></a>

This component has the following requirements:<a name="sm-edge-manager-component-reqs"></a>
+ <a name="sm-req-core-device"></a>A Greengrass core device running on Amazon Linux 2, a Debian-based Linux platform (x86\$164 or Armv8), or Windows (x86\$164). If you don't have one, see [Tutorial: Getting started with AWS IoT Greengrass V2](getting-started.md).
+ <a name="sm-req-python"></a>[Python](https://www.python.org/downloads/) 3.6 or later, including `pip` for your version of Python, installed on your core device.
+ The [Greengrass device role](device-service-role.md) configured with the following: 
  + <a name="sm-req-iam-trust-relationship"></a>A trust relationship that allows `credentials.iot.amazonaws.com` and `sagemaker.amazonaws.com` to assume the role, as shown in the following IAM policy example.

    ```
    { 
      "Version": "2012-10-17",		 	 	 
      "Statement": [ 
        { 
          "Effect": "Allow", 
          "Principal": {
            "Service": "credentials.iot.amazonaws.com"
           }, 
          "Action": "sts:AssumeRole" 
        },
        { 
          "Effect": "Allow", 
          "Principal": {
            "Service": "sagemaker.amazonaws.com"
          }, 
          "Action": "sts:AssumeRole" 
        } 
      ] 
    }
    ```
  + <a name="sm-req-iam-sagemanakeredgedevicefleetpolicy"></a>The [AmazonSageMakerEdgeDeviceFleetPolicy](https://console.aws.amazon.com/iam/home#/policies/arn:aws:iam::aws:policy/service-role/AmazonSageMakerEdgeDeviceFleetPolicy) IAM managed policy.
  + <a name="sm-req-iam-s3-putobject"></a>The `s3:PutObject` action, as shown in the following IAM policy example.

    ```
    {
      "Version": "2012-10-17",		 	 	 
      "Statement": [
        {
          "Action": [
            "s3:PutObject"
          ],
          "Resource": [
            "*"
          ],
          "Effect": "Allow"
        }
      ]
    }
    ```
+ <a name="sm-req-s3-bucket"></a>An Amazon S3 bucket created in the same AWS account and AWS Region as your Greengrass core device. SageMaker AI Edge Manager requires an S3 bucket to create an edge device fleet, and to store sample data from running inference on your device. For information about creating S3 buckets, see [Getting started with Amazon S3](https://docs.aws.amazon.com/AmazonS3/latest/userguide/GetStartedWithS3.html).
+ <a name="sm-req-edge-device-fleet"></a>A SageMaker AI edge device fleet that uses the same AWS IoT role alias as your Greengrass core device. For more information, see [Create an edge device fleet](get-started-with-edge-manager-on-greengrass.md#create-edge-device-fleet-for-greengrass).
+ <a name="sm-req-edge-device"></a>Your Greengrass core device registered as an edge device in your SageMaker AI Edge device fleet. The edge device name must match the AWS IoT thing name for your core device. For more information, see [Register your Greengrass core device](get-started-with-edge-manager-on-greengrass.md#register-greengrass-core-device-in-sme).

### Endpoints and ports
<a name="sagemaker-edge-manager-component-endpoints"></a>

This component must be able to perform outbound requests to the following endpoints and ports, in addition to endpoints and ports required for basic operation. For more information, see [Allow device traffic through a proxy or firewall](allow-device-traffic.md).


| Endpoint | Port | Required | Description | 
| --- | --- | --- | --- | 
|  `edge.sagemaker.region.amazonaws.com`  | 443 | Yes |  Check device registration status and send metrics to SageMaker AI.  | 
|  `*.s3.amazonaws.com`  | 443 | Yes |  Upload capture data to the S3 bucket that you specify. You can replace `*` with the name of each bucket where you upload data.  | 

## Dependencies
<a name="sagemaker-edge-manager-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#sagemaker-edge-manager-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 1.3.5 and 1.3.6 ]

The following table lists the dependencies for version 1.3.5 and 1.3.6 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------
#### [ 1.3.4 ]

The following table lists the dependencies for version 1.3.4 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------
#### [ 1.3.3 ]

The following table lists the dependencies for version 1.3.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------
#### [ 1.3.2 ]

The following table lists the dependencies for version 1.3.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------
#### [ 1.3.1 ]

The following table lists the dependencies for version 1.3.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------
#### [ 1.1.1 - 1.3.0 ]

The following table lists the dependencies for versions 1.1.1 - 1.3.0 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------
#### [ 1.1.0 ]

The following table lists the dependencies for version 1.1.0 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------
#### [ 1.0.3 ]

The following table lists the dependencies for version 1.0.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------
#### [ 1.0.1 and 1.0.2 ]

The following table lists the dependencies for versions 1.0.1 and 1.0.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------
#### [ 1.0.0 ]

The following table lists the dependencies for version 1.0.0 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 
| [Token exchange service](token-exchange-service-component.md) | >=0.0.0 | Hard | 

------

For more information about component dependencies, see the [component recipe reference](component-recipe-reference.md#recipe-reference-component-dependencies).

## Configuration
<a name="sagemaker-edge-manager-component-configuration"></a>

This component provides the following configuration parameters that you can customize when you deploy the component.

**Note**  
This section describes the configuration parameters that you set in the component. For more information about the corresponding SageMaker AI Edge Manager configuration, see [Edge Manager Agent](https://docs.aws.amazon.com/sagemaker/latest/dg/edge-device-fleet-about.html#edge-device-fleet-running-agent) in the *Amazon SageMaker AI Developer Guide*.

`DeviceFleetName`  
The name of the SageMaker AI Edge Manager device fleet that contains your Greengrass core device.   
You must specify a value for this parameter in the configuration update when you deploy this component.

`BucketName`  
The name of the S3 bucket to which you upload captured inference data. The bucket name must contain the string `sagemaker`.   
If you set `CaptureDataDestination` to `Cloud`, or if you set `CaptureDataPeriodicUpload` to `true`, then you must specify a value for this parameter in the configuration update when you deploy this component.  
Capture data is an SageMaker AI feature that you use to upload inference input, inference results, and additional inference data to an S3 bucket or a local directory for future analysis. For more information about using capture data with SageMaker AI Edge Manager, see [Manage Model](https://docs.aws.amazon.com/sagemaker/latest/dg/edge-manage-model.html#edge-manage-model-capturedata) in the *Amazon SageMaker AI Developer Guide*.

`CaptureDataBatchSize`  
(Optional) The size of a batch of capture data requests that the agent handles. This value must be less than the buffer size that you specify in `CaptureDataBufferSize`. We recommend that you don't exceed half the buffer size.  
The agent handles a request batch when the number of requests in the buffer meets the `CaptureDataBatchSize` number, or when the `CaptureDataPushPeriodSeconds` interval elapses, whichever occurs first.  
Default: `10`

`CaptureDataBufferSize`  
(Optional) The maximum number of capture data requests stored in the buffer.  
Default: `30`

`CaptureDataDestination`  
(Optional) The destination where you store captured data. This parameter can have the following values:  
+ `Cloud`—Uploads captured data to the S3 bucket that you specify in `BucketName`.
+ `Disk`—Writes captured data to the component's work directory. 
If you specify `Disk`, you can also choose to periodically upload the captured data to your S3 bucket by setting `CaptureDataPeriodicUpload` to `true`.  
Default: `Cloud`

`CaptureDataPeriodicUpload`  
(Optional) String value that specifies whether to periodically upload captured data. Supported values are `true` and `false`.  
Set this parameter to `true` if you set `CaptureDataDestination` to `Disk`, and you also want the agent to periodically upload the captured data your S3 bucket.  
Default: `false`

`CaptureDataPeriodicUploadPeriodSeconds`  
(Optional) The interval in seconds at which SageMaker AI Edge Manager agent uploads captured data to the S3 bucket. Use this parameter if you set `CaptureDataPeriodicUpload` to `true`.  
Default: `8`

`CaptureDataPushPeriodSeconds`  
(Optional) The interval in seconds at which SageMaker AI Edge Manager agent handles a batch of capture data requests from the buffer.   
The agent handles a request batch when the number of requests in the buffer meets the `CaptureDataBatchSize` number, or when the `CaptureDataPushPeriodSeconds` interval elapses, whichever occurs first.  
Default: `4`

`CaptureDataBase64EmbedLimit`  
(Optional) The maximum size in bytes of captured data that SageMaker AI Edge Manager agent uploads.  
Default: `3072`

`FolderPrefix`  
(Optional) The name of the folder to which the agent writes the captured data. If you set `CaptureDataDestination` to `Disk`, the agent creates the folder in the directory that is specified by `CaptureDataDiskPath`. If you set `CaptureDataDestination` to `Cloud`, or if you set `CaptureDataPeriodicUpload` to `true`, the agent creates the folder in your S3 bucket.   
Default: `sme-capture`

`CaptureDataDiskPath`  
This feature is available in v1.1.0 and later versions of the SageMaker AI Edge Manager component.  
(Optional) The path to the folder to which the agent creates the captured data folder. If you set `CaptureDataDestination` to `Disk`, the agent creates the captured data folder in this directory. If you don't specify this value, the agent creates the captured data folder in the component's work directory. Use the `FolderPrefix` parameter to specify the name of the captured data folder.  
Default: `/greengrass/v2/work/aws.greengrass.SageMakerEdgeManager/capture`

`LocalDataRootPath`  
This feature is available in v1.2.0 and later versions of the SageMaker AI Edge Manager component.  
(Optional) The path where this component stores the following data on the core device:  
+ The local database for runtime data when you set `DbEnable` to `true`.
+ SageMaker AI Neo-compiled models that this component automatically downloads when you set `DeploymentEnable` to `true`.
Default: `/greengrass/v2/work/aws.greengrass.SageMakerEdgeManager`

`DbEnable`  
(Optional) You can enable this component to store runtime data in a local database to preserve the data, in case the component fails or the device loses power.  
This database requires 5 MB of storage on the core device's file system.  
Default: `false`

`DeploymentEnable`  
This feature is available in v1.2.0 and later versions of the SageMaker AI Edge Manager component.  
(Optional) You can enable this component to automatically retrieve SageMaker AI Neo-compiled models from that you upload to Amazon S3. After you upload a new model to Amazon S3, use SageMaker AI Studio or the SageMaker AI API to deploy the new model to this core device. When you enable this feature, you can deploy new models to core devices without needing to create a AWS IoT Greengrass deployment.  
To use this feature, you must set `DbEnable` to `true`. This feature uses the local database to track models that it retrieves from the AWS Cloud.
Default: `false`

`DeploymentPollInterval`  
This feature is available in v1.2.0 and later versions of the SageMaker AI Edge Manager component.  
(Optional) The amount of time (in minutes) between which this component checks for new models to download. This option applies when you set `DeploymentEnable` to `true`.  
Default: `1440` (1 day)

`DLRBackendOptions`  
This feature is available in v1.2.0 and later versions of the SageMaker AI Edge Manager component.  
(Optional) The DLR runtime flags to set in the DLR runtime that this component uses. You can set the following flag:  
+ `TVM_TENSORRT_CACHE_DIR` – Enable TensorRT model caching. Specify an absolute path to an existing folder that has read/write permissions.
+ `TVM_TENSORRT_CACHE_DISK_SIZE_MB` – Assigns the upper limit of the TensorRT model cache folder. When the directory size grows beyond this limit the cached engines that are used the least are deleted. The default value is 512 MB.
For example, you can set this parameter to the following value to enable TensorRT model caching and limit the cache size to 800 MB.  

```
TVM_TENSORRT_CACHE_DIR=/data/secured_folder/trt/cache; TVM_TENSORRT_CACHE_DISK_SIZE_MB=800
```

`SagemakerEdgeLogVerbose`  
(Optional) String value that specifies whether to enable debug logging. Supported values are `true` and `false`.  
Default: `false`

`UnixSocketName`  
(Optional) The location of the SageMaker AI Edge Manager socket file descriptor on the core device.  
Default: `/tmp/aws.greengrass.SageMakerEdgeManager.sock`

**Example: Configuration merge update**  
The following example configuration specifies that the core device is part of the *MyEdgeDeviceFleet* and that the agent writes capture data both to the device and to an S3 bucket. This configuration also enables debug logging.  

```
{
    "DeviceFleetName": "MyEdgeDeviceFleet",
    "BucketName": "amzn-s3-demo-bucket",
    "CaptureDataDestination": "Disk",
    "CaptureDataPeriodicUpload": "true",
    "SagemakerEdgeLogVerbose": "true"    
}
```

## Local log file
<a name="sagemaker-edge-manager-component-log-file"></a>

This component uses the following log file.

------
#### [ Linux ]

```
/greengrass/v2/logs/aws.greengrass.SageMakerEdgeManager.log
```

------
#### [ Windows ]

```
C:\greengrass\v2\logs\aws.greengrass.SageMakerEdgeManager.log
```

------

**To view this component's logs**
+ Run the following command on the core device to view this component's log file in real time. Replace `/greengrass/v2` or *C:\$1greengrass\$1v2* with the path to the AWS IoT Greengrass root folder.

------
#### [ Linux ]

  ```
  sudo tail -f /greengrass/v2/logs/aws.greengrass.SageMakerEdgeManager.log
  ```

------
#### [ Windows (PowerShell) ]

  ```
  Get-Content C:\greengrass\v2\logs\aws.greengrass.SageMakerEdgeManager.log -Tail 10 -Wait
  ```

------

## Changelog
<a name="sagemaker-edge-manager-component-changelog"></a>

The following table describes the changes in each version of the component.


|  **Version**  |  **Changes**  | 
| --- | --- | 
|  1.3.6  |  Version updated for Greengrass nucleus 2.12.5 release.  | 
|  1.3.5  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  1.3.4  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  1.3.3  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  1.3.2  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  1.3.1  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  1.3.0  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/sagemaker-edge-manager-component.html)  | 
|  1.2.0  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/sagemaker-edge-manager-component.html)  | 
|  1.1.1  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  1.1.0  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/sagemaker-edge-manager-component.html)  | 
|  1.0.3  |  Version updated for Greengrass nucleus version 2.4.0 release.  | 
|  1.0.2  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/sagemaker-edge-manager-component.html)  | 
|  1.0.1  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  1.0.0  |  Initial version.  | 

# DLR image classification
<a name="dlr-image-classification-component"></a>

The DLR image classification component (`aws.greengrass.DLRImageClassification`) contains sample inference code to perform image classification inference using [Deep Learning Runtime](https://github.com/neo-ai/neo-ai-dlr) and resnet-50 models. This component uses the variant [DLR image classification model store](dlr-image-classification-model-store-component.md) and the [DLR runtime](dlr-component.md) components as dependencies to download DLR and the sample models. 

To use this inference component with a custom-trained DLR model, [create a custom version](ml-customization.md#override-public-model-store) of the dependent model store component. To use your own custom inference code, you can use the recipe of this component as a template to [create a custom inference component](ml-customization.md#create-inference-component).

**Topics**
+ [Versions](#dlr-image-classification-component-versions)
+ [Type](#dlr-image-classification-component-type)
+ [Operating system](#dlr-image-classification-component-os-support)
+ [Requirements](#dlr-image-classification-component-requirements)
+ [Dependencies](#dlr-image-classification-component-dependencies)
+ [Configuration](#dlr-image-classification-component-configuration)
+ [Local log file](#dlr-image-classification-component-log-file)
+ [Changelog](#dlr-image-classification-component-changelog)

## Versions
<a name="dlr-image-classification-component-versions"></a>

This component has the following versions:
+ 2.1.x
+ 2.0.x

## Type
<a name="dlr-image-classification-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="dlr-image-classification-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="dlr-image-classification-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

## Dependencies
<a name="dlr-image-classification-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#dlr-image-classification-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 2.1.13 and 2.1.14 ]

The following table lists the dependencies for version 2.1.13 and 2.1.14 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.12 ]

The following table lists the dependencies for version 2.1.12 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.11 ]

The following table lists the dependencies for version 2.1.11 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.10 ]

The following table lists the dependencies for version 2.1.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.9 ]

The following table lists the dependencies for version 2.1.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.8 ]

The following table lists the dependencies for version 2.1.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.7 ]

The following table lists the dependencies for version 2.1.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.6 ]

The following table lists the dependencies for version 2.1.6 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.4 - 2.1.5 ]

The following table lists the dependencies for versions 2.1.4 to 2.1.5 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.3 ]

The following table lists the dependencies for version 2.1.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.2 ]

The following table lists the dependencies for version 2.1.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.1 ]

The following table lists the dependencies for version 2.1.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 
| [DLR image classification model store](dlr-image-classification-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.0.x ]

The following table lists the dependencies for version 2.0.x of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | \$12.0.0 | Soft | 
| DLR image classification model store | \$12.0.0 | Hard | 
| DLR | \$11.3.0 | Soft | 

------

## Configuration
<a name="dlr-image-classification-component-configuration"></a>

This component provides the following configuration parameters that you can customize when you deploy the component.

------
#### [ 2.1.x ]

`accessControl`  
<a name="ml-config-accesscontrol-desc"></a>(Optional) The object that contains the [authorization policy](interprocess-communication.md#ipc-authorization-policies) that allows the component to publish messages to the default notifications topic.   
Default:   

```
{
   "aws.greengrass.ipc.mqttproxy": {
      "aws.greengrass.DLRImageClassification:mqttproxy:1": {
         "policyDescription": "Allows access to publish via topic ml/dlr/image-classification.",
         "operations": [
            "aws.greengrass#PublishToIoTCore"
         ],
         "resources": [
            "ml/dlr/image-classification"
         ]
      }
   }
}
```

`PublishResultsOnTopic`  
<a name="ml-config-publishresultsontopic-desc"></a>(Optional) The topic on which you want to publish the inference results. If you modify this value, then you must also modify the value of `resources` in the `accessControl` parameter to match your custom topic name.  
Default: `ml/dlr/image-classification`

`Accelerator`  <a name="ml-config-accelerator"></a>
The accelerator that you want to use. Supported values are `cpu` and `gpu`.  
The sample models in the dependent model component support only CPU acceleration. To use GPU acceleration with a different custom model, [create a custom model component](ml-customization.md#override-public-model-store) to override the public model component.  
Default: `cpu`

`ImageDirectory`  
<a name="ml-config-imagedirectory-desc"></a>(Optional) The path of the folder on the device where inference components read images. You can modify this value to any location on your device to which you have read/write access.  
<a name="ml-config-imagedirectory-img-default"></a>Default: `/greengrass/v2/packages/artifacts-unarchived/component-name/image_classification/sample_images/`  
If you set the value of `UseCamera` to `true`, then this configuration parameter is ignored. 

`ImageName`  
<a name="ml-config-imagename-desc"></a>(Optional) The name of the image that the inference component uses as an input to a make prediction. The component looks for the image in the folder specified in `ImageDirectory`. By default, the component uses the sample image in the default image directory. AWS IoT Greengrass supports the following image formats: `jpeg`, `jpg`, `png`, and `npy`.   
<a name="ml-config-imagename-img-default"></a>Default: `cat.jpeg`  
If you set the value of `UseCamera` to `true`, then this configuration parameter is ignored. 

`InferenceInterval`  <a name="ml-config-inferenceinterval"></a>
(Optional) The time in seconds between each prediction made by the inference code. The sample inference code runs indefinitely and repeats its predictions at the specified time interval. For example, you can change this to a shorter interval if you want to use images taken by a camera for real-time prediction.  
Default: `3600`

`ModelResourceKey`  <a name="ml-config-modelresourcekey"></a>
<a name="ml-config-modelresourcekey-desc"></a>(Optional) The models that are used in the dependent public model component. Modify this parameter only if you override the public model component with a custom component.   
Default:  

```
{
    "armv7l": "DLR-resnet50-armv7l-cpu-ImageClassification",
    "aarch64": "DLR-resnet50-aarch64-cpu-ImageClassification",
    "x86_64": "DLR-resnet50-x86_64-cpu-ImageClassification",
    "windows": "DLR-resnet50-win-cpu-ImageClassification"
}
```

`UseCamera`  <a name="ml-config-usecamera"></a>
(Optional) String value that defines whether to use images from a camera connected to the Greengrass core device. Supported values are `true` and `false`.  
When you set this value to `true`, the sample inference code accesses the camera on your device and runs inference locally on the captured image. The values of the `ImageName` and `ImageDirectory` parameters are ignored. Make sure that the user running this component has read/write access to the location where the camera stores captured images.  
Default: `false`  
When you view the recipe of this component, the `UseCamera` configuration parameter doesn't appear in the default configuration. However, you can modify the value of this parameter in a [configuration merge update](update-component-configurations.md) when you deploy the component.   
When you set `UseCamera` to `true`, you must also create a symlink to enable the inference component to access your camera from the virtual environment that is created by the runtime component. For more information about using a camera with the sample inference components, see [Update component configurations](ml-tutorial-image-classification-camera.md).

------
#### [ 2.0.x ]

`MLRootPath`  <a name="ml-config-mlrootpath"></a>
<a name="ml-config-mlrootpath-desc"></a>(Optional) The path of the folder on Linux core devices where inference components read images and write inference results. You can modify this value to any location on your device to which the user running this component has read/write access.  
<a name="ml-config-mlrootpath-default-dlr"></a>Default: `/greengrass/v2/work/variant.DLR/greengrass_ml`  
<a name="ml-config-mlrootpath-default-tfl"></a>Default: `/greengrass/v2/work/variant.TensorFlowLite/greengrass_ml`

`Accelerator`  <a name="ml-config-accelerator"></a>
The accelerator that you want to use. Supported values are `cpu` and `gpu`.  
The sample models in the dependent model component support only CPU acceleration. To use GPU acceleration with a different custom model, [create a custom model component](ml-customization.md#override-public-model-store) to override the public model component.  
Default: `cpu`

`ImageName`  
<a name="ml-config-imagename-desc-dlr-1.3.0"></a>(Optional) The name of the image that the inference component uses as an input to a make prediction. The component looks for the image in the folder specified in `ImageDirectory`. The default location is `MLRootPath/images`. AWS IoT Greengrass supports the following image formats: `jpeg`, `jpg`, `png`, and `npy`.   
<a name="ml-config-imagename-img-default"></a>Default: `cat.jpeg`

`InferenceInterval`  <a name="ml-config-inferenceinterval"></a>
(Optional) The time in seconds between each prediction made by the inference code. The sample inference code runs indefinitely and repeats its predictions at the specified time interval. For example, you can change this to a shorter interval if you want to use images taken by a camera for real-time prediction.  
Default: `3600`

`ModelResourceKey`  <a name="ml-config-modelresourcekey"></a>
<a name="ml-config-modelresourcekey-desc"></a>(Optional) The models that are used in the dependent public model component. Modify this parameter only if you override the public model component with a custom component.   
Default:  

```
armv7l: "DLR-resnet50-armv7l-cpu-ImageClassification"
x86_64: "DLR-resnet50-x86_64-cpu-ImageClassification"
```

------

## Local log file
<a name="dlr-image-classification-component-log-file"></a>

This component uses the following log file.

------
#### [ Linux ]

```
/greengrass/v2/logs/aws.greengrass.DLRImageClassification.log
```

------
#### [ Windows ]

```
C:\greengrass\v2\logs\aws.greengrass.DLRImageClassification.log
```

------

**To view this component's logs**
+ Run the following command on the core device to view this component's log file in real time. Replace `/greengrass/v2` or *C:\$1greengrass\$1v2* with the path to the AWS IoT Greengrass root folder.

------
#### [ Linux ]

  ```
  sudo tail -f /greengrass/v2/logs/aws.greengrass.DLRImageClassification.log
  ```

------
#### [ Windows (PowerShell) ]

  ```
  Get-Content C:\greengrass\v2\logs\aws.greengrass.DLRImageClassification.log -Tail 10 -Wait
  ```

------

## Changelog
<a name="dlr-image-classification-component-changelog"></a>

The following table describes the changes in each version of the component.


|  **Version**  |  **Changes**  | 
| --- | --- | 
|  2.1.14  | Version updated for Greengrass nucleus 2.12.5 release. | 
|  2.1.13  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  2.1.12  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  2.1.11  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  2.1.10  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  2.1.9  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  2.1.8  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  2.1.7  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  2.1.6  |  Version updated for Greengrass nucleus version 2.5.0 release.  | 
|  2.1.5  |  Component released in all AWS Regions.  | 
|  2.1.4  |  Version updated for Greengrass nucleus version 2.4.0 release. This version isn't available in Europe (London) (`eu-west-2`).  | 
|  2.1.3  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  2.1.2  |  Version updated for Greengrass nucleus version 2.2.0 release.  | 
|  2.1.1  |  <a name="changelog-dlr-image-classification-2.1.1"></a>[\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-image-classification-component.html)  | 
|  2.0.4  |  Initial version.  | 

# DLR object detection
<a name="dlr-object-detection-component"></a>

The DLR object detection component (`aws.greengrass.DLRObjectDetection`) contains sample inference code to perform object detection inference using [Deep Learning Runtime](https://github.com/neo-ai/neo-ai-dlr) and sample pre-trained models. This component uses the variant [DLR object detection model store](dlr-object-detection-model-store-component.md) and the [DLR runtime](dlr-component.md) components as dependencies to download DLR and the sample models. 

To use this inference component with a custom-trained DLR model, [create a custom version](ml-customization.md#override-public-model-store) of the dependent model store component. To use your own custom inference code, you can use the recipe of this component as a template to [create a custom inference component](ml-customization.md#create-inference-component).

**Topics**
+ [Versions](#dlr-object-detection-component-versions)
+ [Type](#dlr-object-detection-component-type)
+ [Operating system](#dlr-object-detection-component-os-support)
+ [Requirements](#dlr-object-detection-component-requirements)
+ [Dependencies](#dlr-object-detection-component-dependencies)
+ [Configuration](#dlr-object-detection-component-configuration)
+ [Local log file](#dlr-object-detection-component-log-file)
+ [Changelog](#dlr-object-detection-component-changelog)

## Versions
<a name="dlr-object-detection-component-versions"></a>

This component has the following versions:
+ 2.1.x
+ 2.0.x

## Type
<a name="dlr-object-detection-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="dlr-object-detection-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="dlr-object-detection-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

## Dependencies
<a name="dlr-object-detection-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#dlr-object-detection-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 2.1.13 and 2.1.14 ]

The following table lists the dependencies for version 2.1.13 and 2.1.14 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.12 ]

The following table lists the dependencies for version 2.1.12 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.11 ]

The following table lists the dependencies for version 2.1.11 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.10 ]

The following table lists the dependencies for version 2.1.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.9 ]

The following table lists the dependencies for version 2.1.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.8 ]

The following table lists the dependencies for version 2.1.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.7 ]

The following table lists the dependencies for version 2.1.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.6 ]

The following table lists the dependencies for version 2.1.6 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.4 - 2.1.5 ]

The following table lists the dependencies for versions 2.1.4 to 2.1.5 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.3 ]

The following table lists the dependencies for version 2.1.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.2 ]

The following table lists the dependencies for version 2.1.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.1.1 ]

The following table lists the dependencies for version 2.1.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 
| [DLR object detection model store](dlr-object-detection-model-store-component.md) | \$12.1.0 | Hard | 
| [DLR](dlr-component.md) | \$11.6.0 | Hard | 

------
#### [ 2.0.x ]

The following table lists the dependencies for version 2.0.x of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | \$12.0.0 | Soft | 
| DLR object detection model store | \$12.0.0 | Hard | 
| DLR | \$11.3.0 | Soft | 

------

## Configuration
<a name="dlr-object-detection-component-configuration"></a>

This component provides the following configuration parameters that you can customize when you deploy the component.

------
#### [ 2.1.x ]

`accessControl`  
<a name="ml-config-accesscontrol-desc"></a>(Optional) The object that contains the [authorization policy](interprocess-communication.md#ipc-authorization-policies) that allows the component to publish messages to the default notifications topic.   
Default:   

```
{
   "aws.greengrass.ipc.mqttproxy": {
      "aws.greengrass.DLRObjectDetection:mqttproxy:1": {
         "policyDescription": "Allows access to publish via topic ml/dlr/object-detection.",
         "operations": [
            "aws.greengrass#PublishToIoTCore"
         ],
         "resources": [
            "ml/dlr/object-detection"
         ]
      }
   }
}
```

`PublishResultsOnTopic`  
<a name="ml-config-publishresultsontopic-desc"></a>(Optional) The topic on which you want to publish the inference results. If you modify this value, then you must also modify the value of `resources` in the `accessControl` parameter to match your custom topic name.  
Default: `ml/dlr/object-detection`

`Accelerator`  <a name="ml-config-accelerator"></a>
The accelerator that you want to use. Supported values are `cpu` and `gpu`.  
The sample models in the dependent model component support only CPU acceleration. To use GPU acceleration with a different custom model, [create a custom model component](ml-customization.md#override-public-model-store) to override the public model component.  
Default: `cpu`

`ImageDirectory`  
<a name="ml-config-imagedirectory-desc"></a>(Optional) The path of the folder on the device where inference components read images. You can modify this value to any location on your device to which you have read/write access.  
<a name="ml-config-imagedirectory-obj-default"></a>Default: `/greengrass/v2/packages/artifacts-unarchived/component-name/object_detection/sample_images/`  
If you set the value of `UseCamera` to `true`, then this configuration parameter is ignored. 

`ImageName`  
<a name="ml-config-imagename-desc"></a>(Optional) The name of the image that the inference component uses as an input to a make prediction. The component looks for the image in the folder specified in `ImageDirectory`. By default, the component uses the sample image in the default image directory. AWS IoT Greengrass supports the following image formats: `jpeg`, `jpg`, `png`, and `npy`.   
<a name="ml-config-imagename-obj-default"></a>Default: `objects.jpg`  
If you set the value of `UseCamera` to `true`, then this configuration parameter is ignored. 

`InferenceInterval`  <a name="ml-config-inferenceinterval"></a>
(Optional) The time in seconds between each prediction made by the inference code. The sample inference code runs indefinitely and repeats its predictions at the specified time interval. For example, you can change this to a shorter interval if you want to use images taken by a camera for real-time prediction.  
Default: `3600`

`ModelResourceKey`  <a name="ml-config-modelresourcekey"></a>
<a name="ml-config-modelresourcekey-desc"></a>(Optional) The models that are used in the dependent public model component. Modify this parameter only if you override the public model component with a custom component.   
Default:  

```
{
    "armv7l": "DLR-yolo3-armv7l-cpu-ObjectDetection",
    "aarch64": "DLR-yolo3-aarch64-gpu-ObjectDetection",
    "x86_64": "DLR-yolo3-x86_64-cpu-ObjectDetection",
    "windows": "DLR-resnet50-win-cpu-ObjectDetection"
}
```

`UseCamera`  <a name="ml-config-usecamera"></a>
(Optional) String value that defines whether to use images from a camera connected to the Greengrass core device. Supported values are `true` and `false`.  
When you set this value to `true`, the sample inference code accesses the camera on your device and runs inference locally on the captured image. The values of the `ImageName` and `ImageDirectory` parameters are ignored. Make sure that the user running this component has read/write access to the location where the camera stores captured images.  
Default: `false`  
When you view the recipe of this component, the `UseCamera` configuration parameter doesn't appear in the default configuration. However, you can modify the value of this parameter in a [configuration merge update](update-component-configurations.md) when you deploy the component.   
When you set `UseCamera` to `true`, you must also create a symlink to enable the inference component to access your camera from the virtual environment that is created by the runtime component. For more information about using a camera with the sample inference components, see [Update component configurations](ml-tutorial-image-classification-camera.md).

------
#### [ 2.0.x ]

`MLRootPath`  <a name="ml-config-mlrootpath"></a>
<a name="ml-config-mlrootpath-desc"></a>(Optional) The path of the folder on Linux core devices where inference components read images and write inference results. You can modify this value to any location on your device to which the user running this component has read/write access.  
<a name="ml-config-mlrootpath-default-dlr"></a>Default: `/greengrass/v2/work/variant.DLR/greengrass_ml`  
<a name="ml-config-mlrootpath-default-tfl"></a>Default: `/greengrass/v2/work/variant.TensorFlowLite/greengrass_ml`

`Accelerator`  
Do not modify. Currently, the only supported value for the accelerator is `cpu`, because the models in the dependent model components are compiled only for the CPU accelerator.

`ImageName`  
<a name="ml-config-imagename-desc-dlr-1.3.0"></a>(Optional) The name of the image that the inference component uses as an input to a make prediction. The component looks for the image in the folder specified in `ImageDirectory`. The default location is `MLRootPath/images`. AWS IoT Greengrass supports the following image formats: `jpeg`, `jpg`, `png`, and `npy`.   
<a name="ml-config-imagename-obj-default"></a>Default: `objects.jpg`

`InferenceInterval`  <a name="ml-config-inferenceinterval"></a>
(Optional) The time in seconds between each prediction made by the inference code. The sample inference code runs indefinitely and repeats its predictions at the specified time interval. For example, you can change this to a shorter interval if you want to use images taken by a camera for real-time prediction.  
Default: `3600`

`ModelResourceKey`  <a name="ml-config-modelresourcekey"></a>
<a name="ml-config-modelresourcekey-desc"></a>(Optional) The models that are used in the dependent public model component. Modify this parameter only if you override the public model component with a custom component.   
Default:  

```
{
    armv7l: "DLR-yolo3-armv7l-cpu-ObjectDetection",
    x86_64: "DLR-yolo3-x86_64-cpu-ObjectDetection"
}
```

------

## Local log file
<a name="dlr-object-detection-component-log-file"></a>

This component uses the following log file.

------
#### [ Linux ]

```
/greengrass/v2/logs/aws.greengrass.DLRObjectDetection.log
```

------
#### [ Windows ]

```
C:\greengrass\v2\logs\aws.greengrass.DLRObjectDetection.log
```

------

**To view this component's logs**
+ Run the following command on the core device to view this component's log file in real time. Replace `/greengrass/v2` or *C:\$1greengrass\$1v2* with the path to the AWS IoT Greengrass root folder.

------
#### [ Linux ]

  ```
  sudo tail -f /greengrass/v2/logs/aws.greengrass.DLRObjectDetection.log
  ```

------
#### [ Windows (PowerShell) ]

  ```
  Get-Content C:\greengrass\v2\logs\aws.greengrass.DLRObjectDetection.log -Tail 10 -Wait
  ```

------

## Changelog
<a name="dlr-object-detection-component-changelog"></a>

The following table describes the changes in each version of the component.


|  **Version**  |  **Changes**  | 
| --- | --- | 
|  2.1.14  | Version updated for Greengrass nucleus 2.12.5 release. | 
|  2.1.13  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  2.1.12  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  2.1.11  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  2.1.10  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  2.1.9  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  2.1.8  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  2.1.7  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  2.1.6  |  Version updated for Greengrass nucleus version 2.5.0 release.  | 
|  2.1.5  |  Component released in all AWS Regions.  | 
|  2.1.4  |  Version updated for Greengrass nucleus version 2.4.0 release. This version isn't available in Europe (London) (`eu-west-2`).  | 
|  2.1.3  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  2.1.2  |  <a name="changelog-dlr-object-detection-2.1.2"></a>[\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-object-detection-component.html)  | 
|  2.1.1  |  <a name="changelog-dlr-object-detection-2.1.1"></a>[\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-object-detection-component.html)  | 
|  2.0.4  |  Initial version.  | 

# DLR image classification model store
<a name="dlr-image-classification-model-store-component"></a>

The DLR image classification model store is a machine learning model component that contains pre-trained ResNet-50 models as Greengrass artifacts. The pre-trained models used in this component are fetched from the [GluonCV Model Zoo](https://cv.gluon.ai/model_zoo/index.html) and are compiled using SageMaker AI Neo [Deep Learning Runtime](https://github.com/neo-ai/neo-ai-dlr).

The [DLR image classification](dlr-image-classification-component.md) inference component uses this component as a dependency for the model source. To use a custom-trained DLR model, [create a custom version](ml-customization.md#override-public-model-store) of this model component, and include your custom model as a component artifact. You can use the recipe of this component as a template to create custom model components. 

**Note**  
The name of the DLR image classification model store component varies depending on its version. The component name for version 2.1.x and later versions is `variant.DLR.ImageClassification.ModelStore`. The component name for version 2.0.x is `variant.ImageClassification.ModelStore`.

**Topics**
+ [Versions](#dlr-image-classification-model-store-component-versions)
+ [Type](#dlr-image-classification-model-store-component-type)
+ [Operating system](#dlr-image-classification-model-store-component-os-support)
+ [Requirements](#dlr-image-classification-model-store-component-requirements)
+ [Dependencies](#dlr-image-classification-model-store-component-dependencies)
+ [Configuration](#dlr-image-classification-model-store-component-configuration)
+ [Local log file](#dlr-image-classification-model-store-component-log-file)
+ [Changelog](#dlr-image-classification-model-store-component-changelog)

## Versions
<a name="dlr-image-classification-model-store-component-versions"></a>

This component has the following versions:
+ 2.1.x (`variant.DLR.ImageClassification.ModelStore`) 
+ 2.0.x (`variant.ImageClassification.ModelStore`) 

## Type
<a name="dlr-image-classification-model-store-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="dlr-image-classification-model-store-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="dlr-image-classification-model-store-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

## Dependencies
<a name="dlr-image-classification-model-store-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#dlr-image-classification-model-store-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 2.1.12 - 2.1.14 ]

The following table lists the dependencies for version 2.1.12 and 2.1.13 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 

------
#### [ 2.1.11 ]

The following table lists the dependencies for version 2.1.11 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 

------
#### [ 2.1.10 ]

The following table lists the dependencies for version 2.1.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 

------
#### [ 2.1.9 ]

The following table lists the dependencies for version 2.1.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 

------
#### [ 2.1.8 ]

The following table lists the dependencies for version 2.1.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 

------
#### [ 2.1.7 ]

The following table lists the dependencies for version 2.1.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 

------
#### [ 2.1.6 ]

The following table lists the dependencies for version 2.1.6 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 

------
#### [ 2.1.5 ]

The following table lists the dependencies for version 2.1.5 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 

------
#### [ 2.1.4 ]

The following table lists the dependencies for version 2.1.4 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 

------
#### [ 2.1.3 ]

The following table lists the dependencies for version 2.1.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 

------
#### [ 2.1.2 ]

The following table lists the dependencies for version 2.1.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 

------
#### [ 2.1.1 ]

The following table lists the dependencies for version 2.1.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 

------
#### [ 2.0.x ]

The following table lists the dependencies for version 2.0.x of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | \$12.0.0 | Soft | 

------

## Configuration
<a name="dlr-image-classification-model-store-component-configuration"></a>

This component doesn't have any configuration parameters.

## Local log file
<a name="dlr-image-classification-model-store-component-log-file"></a>

This component doesn't output logs.

## Changelog
<a name="dlr-image-classification-model-store-component-changelog"></a>

The following table describes the changes in each version of the component.


|  **Version**  |  **Changes**  | 
| --- | --- | 
|  2.1.13  | Version updated for Greengrass nucleus 2.12.5 release. | 
|  2.1.12  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  2.1.11  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  2.1.10  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  2.1.9  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  2.1.8  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  2.1.7  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  2.1.6  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  2.1.5  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-image-classification-model-store-component.html)  | 
|  2.1.4  |  Version updated for Greengrass nucleus version 2.4.0 release.  | 
|  2.1.3  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  2.1.2  |  Version updated for Greengrass nucleus version 2.2.0 release.  | 
|  2.1.1  |  <a name="changelog-dlr-image-classification-model-store-2.1.1"></a>[\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-image-classification-model-store-component.html)  | 
|  2.0.4  |  Initial version.  | 

# DLR object detection model store
<a name="dlr-object-detection-model-store-component"></a>

The DLR object detection model store is a machine learning model component that contains pre-trained YOLOv3 models as Greengrass artifacts. The sample models used in this component are fetched from the [GluonCV Model Zoo](https://cv.gluon.ai/model_zoo/index.html) and compiled using SageMaker AI Neo [Deep Learning Runtime](https://github.com/neo-ai/neo-ai-dlr).

The [DLR object detection](dlr-object-detection-component.md) inference component uses this component as a dependency for the model source. To use a custom-trained DLR model, [create a custom version](ml-customization.md#override-public-model-store) of this model component, and include your custom model as a component artifact. You can use the recipe of this component as a template to create custom model components. 

**Note**  
The name of the DLR object detection model store component varies depending on its version. The component name for version 2.1.x and later versions is `variant.DLR.ObjectDetection.ModelStore`. The component name for version 2.0.x is `variant.ObjectDetection.ModelStore`.

**Topics**
+ [Versions](#dlr-object-detection-model-store-component-versions)
+ [Type](#dlr-object-detection-model-store-component-type)
+ [Operating system](#dlr-object-detection-model-store-component-os-support)
+ [Requirements](#dlr-object-detection-model-store-component-requirements)
+ [Dependencies](#dlr-object-detection-model-store-component-dependencies)
+ [Configuration](#dlr-object-detection-model-store-component-configuration)
+ [Local log file](#dlr-object-detection-model-store-component-log-file)
+ [Changelog](#dlr-object-detection-model-store-component-changelog)

## Versions
<a name="dlr-object-detection-model-store-component-versions"></a>

This component has the following versions:
+ 2.1.x 
+ 2.0.x

## Type
<a name="dlr-object-detection-model-store-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="dlr-object-detection-model-store-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="dlr-object-detection-model-store-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

## Dependencies
<a name="dlr-object-detection-model-store-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#dlr-object-detection-model-store-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 2.1.13 and 2.1.14 ]

The following table lists the dependencies for version 2.1.13 and 2.1.14 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 

------
#### [ 2.1.12 ]

The following table lists the dependencies for version 2.1.12 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 

------
#### [ 2.1.11 ]

The following table lists the dependencies for version 2.1.11 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 

------
#### [ 2.1.10 ]

The following table lists the dependencies for version 2.1.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 

------
#### [ 2.1.9 ]

The following table lists the dependencies for version 2.1.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 

------
#### [ 2.1.8 ]

The following table lists the dependencies for version 2.1.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 

------
#### [ 2.1.7 ]

The following table lists the dependencies for version 2.1.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 

------
#### [ 2.1.5 and 2.1.6 ]

The following table lists the dependencies for versions 2.1.5 and 2.1.6 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 

------
#### [ 2.1.4 ]

The following table lists the dependencies for version 2.1.4 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 

------
#### [ 2.1.3 ]

The following table lists the dependencies for version 2.1.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 

------
#### [ 2.1.2 ]

The following table lists the dependencies for version 2.1.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 

------
#### [ 2.1.1 ]

The following table lists the dependencies for version 2.1.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 

------
#### [ 2.0.x ]

The following table lists the dependencies for version 2.0.x of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | \$12.0.0 | Soft | 

------

## Configuration
<a name="dlr-object-detection-model-store-component-configuration"></a>

This component doesn't have any configuration parameters.

## Local log file
<a name="dlr-object-detection-model-store-component-log-file"></a>

This component doesn't output logs.

## Changelog
<a name="dlr-object-detection-model-store-component-changelog"></a>

The following table describes the changes in each version of the component.


|  **Version**  |  **Changes**  | 
| --- | --- | 
|  2.1.14  |  Version updated for Greengrass nucleus 2.12.5 release.  | 
|  2.1.13  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  2.1.12  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  2.1.11  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  2.1.10  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  2.1.9  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  2.1.8  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  2.1.7  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  2.1.6  |  Adds a CPU model to fix an issue on Armv8 (AArch64) devices.  | 
|  2.1.5  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-object-detection-model-store-component.html)  | 
|  2.1.4  |  Version updated for Greengrass nucleus version 2.4.0 release.  | 
|  2.1.3  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  2.1.2  |  Version updated for Greengrass nucleus version 2.2.0 release.  | 
|  2.1.1  |  <a name="changelog-dlr-object-detection-model-store-2.1.1"></a>[\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-object-detection-model-store-component.html) | 
|  2.0.4  |  Initial version.  | 

# DLR runtime
<a name="dlr-component"></a>

The DLR runtime component (`variant.DLR`) contains a script that installs [Deep Learning Runtime](https://github.com/neo-ai/neo-ai-dlr) (DLR) and its dependencies in a virtual environment on your device. The [DLR image classification](dlr-image-classification-component.md) and [DLR object detection](dlr-object-detection-component.md) components use this component as a dependency for installing DLR. Component version 1.6.x installs DLR v1.6.0 and component version 1.3.x installs DLR v1.3.0. 

To use a different runtime, you can use the recipe of this component as a template to [create a custom machine learning component](ml-customization.md). 

**Topics**
+ [Versions](#dlr-component-versions)
+ [Type](#dlr-component-type)
+ [Operating system](#dlr-component-os-support)
+ [Requirements](#dlr-component-requirements)
+ [Dependencies](#dlr-component-dependencies)
+ [Configuration](#dlr-component-configuration)
+ [Usage](#dlr-component-usage)
+ [Local log file](#dlr-component-log-file)
+ [Changelog](#dlr-component-changelog)

## Versions
<a name="dlr-component-versions"></a>

This component has the following versions:
+ 1.6.x
+ 1.3.x

## Type
<a name="dlr-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="dlr-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="dlr-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

### Endpoints and ports
<a name="dlr-component-endpoints"></a>

By default, this component uses an installer script to install packages using the `apt`, `yum`, `brew`, and `pip` commands, depending on what platform the core device uses. This component must be able to perform outbound requests to various package indexes and repositories to run the installer script. To allow this component's outbound traffic through a proxy or firewall, you must identify the endpoints for the package indexes and repositories where your core device connects to install.

Consider the following when you identify endpoints required for this component's install script:
+ The endpoints depend on the core device's platform. For example, a core device that runs Ubuntu uses `apt` rather than `yum` or `brew`. Additionally, devices that use the same package index might have different source lists, so they might retrieve packages from different repositories.
+ The endpoints might differ between multiple devices that use the same package index, because each device has its own source lists that define where to retrieve packages.
+ The endpoints might change over time. Each package index provides the URLs of the repositories where you download packages, and the owner of a package can change what URLs the package index provides.

For more information about the dependencies that this component installs, and how to disable the installer script, see the [UseInstaller](#dlr-component-config-useinstaller-term) configuration parameter.

For more information about endpoints and ports required for basic operation, see [Allow device traffic through a proxy or firewall](allow-device-traffic.md).

## Dependencies
<a name="dlr-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#dlr-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 1.6.11 - 1.6.16 ]

The following table lists the dependencies for versions 1.6.11 to 1.6.16 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <3.0.0 | Soft | 

------
#### [ 1.6.10 ]

The following table lists the dependencies for version 1.6.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 

------
#### [ 1.6.9 ]

The following table lists the dependencies for version 1.6.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 

------
#### [ 1.6.8 ]

The following table lists the dependencies for version 1.6.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 

------
#### [ 1.6.6 and 1.6.7 ]

The following table lists the dependencies for versions 1.6.6 and 1.6.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 

------
#### [ 1.6.4 and 1.6.5 ]

The following table lists the dependencies for versions 1.6.4 and 1.6.5 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 

------
#### [ 1.6.3 ]

The following table lists the dependencies for version 1.6.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 

------
#### [ 1.6.2 ]

The following table lists the dependencies for version 1.6.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 

------
#### [ 1.6.1 ]

The following table lists the dependencies for version 1.6.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 

------
#### [ 1.3.x ]

The following table lists the dependencies for version 1.3.x of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | \$12.0.0 | Soft | 

------

For more information about component dependencies, see the [component recipe reference](component-recipe-reference.md#recipe-reference-component-dependencies).

## Configuration
<a name="dlr-component-configuration"></a>

This component provides the following configuration parameters that you can customize when you deploy the component.

`MLRootPath`  
<a name="ml-config-mlrootpath-desc"></a>(Optional) The path of the folder on Linux core devices where inference components read images and write inference results. You can modify this value to any location on your device to which the user running this component has read/write access.  
<a name="ml-config-mlrootpath-default-dlr"></a>Default: `/greengrass/v2/work/variant.DLR/greengrass_ml`

`WindowsMLRootPath`  
This feature is available in v1.6.6 and later of this component.  
<a name="ml-config-windowsmlrootpath-desc"></a>(Optional) The path of the folder on Windows core device where inference components read images and write inference results. You can modify this value to any location on your device to which the user running this component has read/write access.  
<a name="ml-config-windowsmlrootpath-default-dlr"></a>Default: `C:\greengrass\v2\\work\\variant.DLR\\greengrass_ml`

  `UseInstaller`   
<a name="ml-config-useinstaller-desc-dlr"></a>(Optional) String value that defines whether to use the installer script in this component to install DLR and its dependencies. Supported values are `true` and `false`.   <a name="ml-config-useinstaller-libraries-dlr"></a>

Set this value to `false` if you want to use a custom script for DLR installation, or if you want to include runtime dependencies in a pre-built Linux image. To use this component with the AWS-provided DLR inference components, install the following libraries, including any dependencies, and make them available to the system user, such as `ggc_user`, that runs the ML components.
+ [Python](https://www.python.org/downloads/) 3.7 or later, including `pip` for your version of Python.
+ [Deep Learning Runtime](https://github.com/neo-ai/neo-ai-dlr) v1.6.0
+ [NumPy](https://numpy.org/install/).
+ [OpenCV-Python](https://pypi.org/project/opencv-python/).
+ [AWS IoT Device SDK v2 for Python](https://github.com/aws/aws-iot-device-sdk-python-v2).
+ [AWS Common Runtime (CRT) Python](https://github.com/awslabs/aws-crt-python).
+ [Picamera](https://picamera.readthedocs.io/en/release-1.13/) (for Raspberry Pi devices only).
+ [`awscam` module](https://docs.aws.amazon.com/deeplens/latest/dg/deeplens-library-awscam-module.html) (for AWS DeepLens devices).
+ libGL (for Linux devices)
<a name="ml-config-useinstaller-default"></a>Default: `true`

## Usage
<a name="dlr-component-usage"></a>

Use this component with the `UseInstaller` configuration parameter set to `true` to install DLR and its dependencies on your device. The component sets up a virtual environment on your device that includes the OpenCV and NumPy libraries that are required for DLR. 

**Note**  <a name="ml-installer-component-usage-note"></a>
The installer script in this component also installs the latest versions of additional system libraries that are required to configure the virtual environment on your device and to use the installed machine learning framework. This might upgrade the existing system libraries on your device. Review the following table for the list of libraries that this component installs for each supported operating system. If you want to customize this installation process, set the `UseInstaller` configuration parameter to `false`, and develop your own installer script.


| Platform | Libraries installed on the device system | Libraries installed in the virtual environment | 
| --- | --- | --- | 
| Armv7l | build-essential, cmake, ca-certificates, git | setuptools, wheel | 
| Amazon Linux 2 | mesa-libGL | None | 
| Ubuntu | wget | None | 

When you deploy your inference component, this runtime component first verifies if your device already has DLR and its dependencies installed, and if not, then it installs them for you. 

## Local log file
<a name="dlr-component-log-file"></a>

This component uses the following log file.

------
#### [ Linux ]

```
/greengrass/v2/logs/variant.DLR.log
```

------
#### [ Windows ]

```
C:\greengrass\v2\logs\variant.DLR.log
```

------

**To view this component's logs**
+ Run the following command on the core device to view this component's log file in real time. Replace `/greengrass/v2` or *C:\$1greengrass\$1v2* with the path to the AWS IoT Greengrass root folder.

------
#### [ Linux ]

  ```
  sudo tail -f /greengrass/v2/logs/variant.DLR.log
  ```

------
#### [ Windows (PowerShell) ]

  ```
  Get-Content C:\greengrass\v2\logs\variant.DLR.log -Tail 10 -Wait
  ```

------

## Changelog
<a name="dlr-component-changelog"></a>

The following table describes the changes in each version of the component.


|  **Version**  |  **Changes**  | 
| --- | --- | 
|  1.6.16  |  Version updated for Greengrass nucleus version 2.12.5.  | 
|  1.6.12  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-component.html)  | 
|  1.6.11  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  1.6.10  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  1.6.9  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  1.6.8  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  1.6.7  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-component.html)  | 
|  1.6.6  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-component.html)  | 
|  1.6.5  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-component.html)  | 
|  1.6.4  |  Version updated for Greengrass nucleus version 2.4.0 release.  | 
|  1.6.3  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  1.6.2  |  Version updated for Greengrass nucleus version 2.2.0 release.  | 
|  1.6.1  |  <a name="changelog-dlr-1.6.1"></a>[\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/dlr-component.html)  | 
|  1.3.2  |  Initial version. Installs DLR v1.3.0.  | 

# TensorFlow Lite image classification
<a name="tensorflow-lite-image-classification-component"></a>

The TensorFlow Lite image classification component (`aws.greengrass.TensorFlowLiteImageClassification`) contains sample inference code to perform image classification inference using the [TensorFlow Lite](https://www.tensorflow.org/lite/guide/python) runtime and a sample pre-trained MobileNet 1.0 quantized model. This component uses the variant [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) and the [TensorFlow Lite runtime](tensorflow-lite-component.md) components as dependencies to download the TensorFlow Lite runtime and the sample model.

To use this inference component with a custom-trained TensorFlow Lite model, [create a custom version](ml-customization.md#override-public-model-store) of the dependent model store component. To use your own custom inference code, you can use the recipe of this component as a template to [create a custom inference component](ml-customization.md#create-inference-component).

**Topics**
+ [Versions](#tensorflow-lite-image-classification-component-versions)
+ [Type](#tensorflow-lite-image-classification-component-type)
+ [Operating system](#tensorflow-lite-image-classification-component-os-support)
+ [Requirements](#tensorflow-lite-image-classification-component-requirements)
+ [Dependencies](#tensorflow-lite-image-classification-component-dependencies)
+ [Configuration](#tensorflow-lite-image-classification-component-configuration)
+ [Local log file](#tensorflow-lite-image-classification-component-log-file)
+ [Changelog](#tensorflow-lite-image-classification-component-changelog)

## Versions
<a name="tensorflow-lite-image-classification-component-versions"></a>

This component has the following versions:
+ 2.1.x

## Type
<a name="tensorflow-lite-image-classification-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="tensorflow-lite-image-classification-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="tensorflow-lite-image-classification-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

## Dependencies
<a name="tensorflow-lite-image-classification-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#tensorflow-lite-image-classification-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 2.1.11 and 2.1.12 ]

The following table lists the dependencies for version 2.1.11 and 2.1.12 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.10 ]

The following table lists the dependencies for version 2.1.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.9 ]

The following table lists the dependencies for version 2.1.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.8 ]

The following table lists the dependencies for version 2.1.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.7 ]

The following table lists the dependencies for version 2.1.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.6 ]

The following table lists the dependencies for version 2.1.6 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.5 ]

The following table lists the dependencies for version 2.1.5 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.4 ]

The following table lists the dependencies for version 2.1.4 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.3 ]

The following table lists the dependencies for version 2.1.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.2 ]

The following table lists the dependencies for version 2.1.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.1 ]

The following table lists the dependencies for version 2.1.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.0 ]

The following table lists the dependencies for version 2.1.0 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------

## Configuration
<a name="tensorflow-lite-image-classification-component-configuration"></a>

This component provides the following configuration parameters that you can customize when you deploy the component.

`accessControl`  
<a name="ml-config-accesscontrol-desc"></a>(Optional) The object that contains the [authorization policy](interprocess-communication.md#ipc-authorization-policies) that allows the component to publish messages to the default notifications topic.   
Default:   

```
{
   "aws.greengrass.ipc.mqttproxy": {
      "aws.greengrass.TensorFlowLiteImageClassification:mqttproxy:1": {
         "policyDescription": "Allows access to publish via topic ml/tflite/image-classification.",
         "operations": [
            "aws.greengrass#PublishToIoTCore"
         ],
         "resources": [
            "ml/tflite/image-classification"
         ]
      }
   }
}
```

`PublishResultsOnTopic`  
<a name="ml-config-publishresultsontopic-desc"></a>(Optional) The topic on which you want to publish the inference results. If you modify this value, then you must also modify the value of `resources` in the `accessControl` parameter to match your custom topic name.  
Default: `ml/tflite/image-classification`

`Accelerator`  <a name="ml-config-accelerator"></a>
The accelerator that you want to use. Supported values are `cpu` and `gpu`.  
The sample models in the dependent model component support only CPU acceleration. To use GPU acceleration with a different custom model, [create a custom model component](ml-customization.md#override-public-model-store) to override the public model component.  
Default: `cpu`

`ImageDirectory`  
<a name="ml-config-imagedirectory-desc"></a>(Optional) The path of the folder on the device where inference components read images. You can modify this value to any location on your device to which you have read/write access.  
<a name="ml-config-imagedirectory-img-default"></a>Default: `/greengrass/v2/packages/artifacts-unarchived/component-name/image_classification/sample_images/`  
If you set the value of `UseCamera` to `true`, then this configuration parameter is ignored. 

`ImageName`  
<a name="ml-config-imagename-desc"></a>(Optional) The name of the image that the inference component uses as an input to a make prediction. The component looks for the image in the folder specified in `ImageDirectory`. By default, the component uses the sample image in the default image directory. AWS IoT Greengrass supports the following image formats: `jpeg`, `jpg`, `png`, and `npy`.   
<a name="ml-config-imagename-img-default"></a>Default: `cat.jpeg`  
If you set the value of `UseCamera` to `true`, then this configuration parameter is ignored. 

`InferenceInterval`  <a name="ml-config-inferenceinterval"></a>
(Optional) The time in seconds between each prediction made by the inference code. The sample inference code runs indefinitely and repeats its predictions at the specified time interval. For example, you can change this to a shorter interval if you want to use images taken by a camera for real-time prediction.  
Default: `3600`

`ModelResourceKey`  <a name="ml-config-modelresourcekey"></a>
<a name="ml-config-modelresourcekey-desc"></a>(Optional) The models that are used in the dependent public model component. Modify this parameter only if you override the public model component with a custom component.   
Default:  

```
{
    "model": "TensorFlowLite-Mobilenet"
}
```

`UseCamera`  <a name="ml-config-usecamera"></a>
(Optional) String value that defines whether to use images from a camera connected to the Greengrass core device. Supported values are `true` and `false`.  
When you set this value to `true`, the sample inference code accesses the camera on your device and runs inference locally on the captured image. The values of the `ImageName` and `ImageDirectory` parameters are ignored. Make sure that the user running this component has read/write access to the location where the camera stores captured images.  
Default: `false`  
When you view the recipe of this component, the `UseCamera` configuration parameter doesn't appear in the default configuration. However, you can modify the value of this parameter in a [configuration merge update](update-component-configurations.md) when you deploy the component.   
When you set `UseCamera` to `true`, you must also create a symlink to enable the inference component to access your camera from the virtual environment that is created by the runtime component. For more information about using a camera with the sample inference components, see [Update component configurations](ml-tutorial-image-classification-camera.md).

## Local log file
<a name="tensorflow-lite-image-classification-component-log-file"></a>

This component uses the following log file.

------
#### [ Linux ]

```
/greengrass/v2/logs/aws.greengrass.TensorFlowLiteImageClassification.log
```

------
#### [ Windows ]

```
C:\greengrass\v2\logs\aws.greengrass.TensorFlowLiteImageClassification.log
```

------

**To view this component's logs**
+ Run the following command on the core device to view this component's log file in real time. Replace `/greengrass/v2` or *C:\$1greengrass\$1v2* with the path to the AWS IoT Greengrass root folder.

------
#### [ Linux ]

  ```
  sudo tail -f /greengrass/v2/logs/aws.greengrass.TensorFlowLiteImageClassification.log
  ```

------
#### [ Windows (PowerShell) ]

  ```
  Get-Content C:\greengrass\v2\logs\aws.greengrass.TensorFlowLiteImageClassification.log -Tail 10 -Wait
  ```

------

## Changelog
<a name="tensorflow-lite-image-classification-component-changelog"></a>

The following table describes the changes in each version of the component.


|  Version  |  Changes  | 
| --- | --- | 
|  2.1.12  |  Version updated for Greengrass nucleus 2.12.5 release.  | 
|  2.1.11  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  2.1.10  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  2.1.9  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  2.1.8  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  2.1.7  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  2.1.6  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  2.1.5  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  2.1.4  |  Version updated for Greengrass nucleus version 2.5.0 release.  | 
|  2.1.3  |  Version updated for Greengrass nucleus version 2.4.0 release.  | 
|  2.1.2  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  2.1.1  |  Version updated for Greengrass nucleus version 2.2.0 release.  | 
|  2.1.0  |  Initial version.  | 

# TensorFlow Lite object detection
<a name="tensorflow-lite-object-detection-component"></a>

The TensorFlow Lite object detection component (`aws.greengrass.TensorFlowLiteObjectDetection`) contains sample inference code to perform object detection inference using [TensorFlow Lite](https://www.tensorflow.org/lite/guide/python) and a sample pre-trained Single Shot Detection (SSD) MobileNet 1.0 model. This component uses the variant [TensorFlow Lite object detection model store](tensorflow-lite-object-detection-model-store-component.md) and the [TensorFlow Lite runtime](tensorflow-lite-component.md) components as dependencies to download TensorFlow Lite and the sample model. 

To use this inference component with a custom-trained TensorFlow Lite model, you can [create a custom version](ml-customization.md#override-public-model-store) of the dependent model store component. To use your own custom inference code, use the recipe of this component as a template to [create a custom inference component](ml-customization.md#create-inference-component).

**Topics**
+ [Versions](#tensorflow-lite-object-detection-component-versions)
+ [Type](#tensorflow-lite-object-detection-component-type)
+ [Operating system](#tensorflow-lite-object-detection-component-os-support)
+ [Requirements](#tensorflow-lite-object-detection-component-requirements)
+ [Dependencies](#tensorflow-lite-object-detection-component-dependencies)
+ [Configuration](#tensorflow-lite-object-detection-component-configuration)
+ [Local log file](#tensorflow-lite-object-detection-component-log-file)
+ [Changelog](#tensorflow-lite-object-detection-component-changelog)

## Versions
<a name="tensorflow-lite-object-detection-component-versions"></a>

This component has the following versions:
+ 2.1.x

## Type
<a name="tensorflow-lite-object-detection-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="tensorflow-lite-object-detection-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="tensorflow-lite-object-detection-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

## Dependencies
<a name="tensorflow-lite-object-detection-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#tensorflow-lite-object-detection-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 2.1.11 and 2.1.12 ]

The following table lists the dependencies for version 2.1.11 and 2.1.12 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.10 ]

The following table lists the dependencies for version 2.1.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.9 ]

The following table lists the dependencies for version 2.1.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.8 ]

The following table lists the dependencies for version 2.1.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.7 ]

The following table lists the dependencies for version 2.1.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.6 ]

The following table lists the dependencies for version 2.1.6 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.5 ]

The following table lists the dependencies for version 2.1.5 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.4 ]

The following table lists the dependencies for version 2.1.4 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.3 ]

The following table lists the dependencies for version 2.1.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.2 ]

The following table lists the dependencies for version 2.1.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.1 ]

The following table lists the dependencies for version 2.1.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------
#### [ 2.1.0 ]

The following table lists the dependencies for version 2.1.0 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 
| [TensorFlow Lite image classification model store](tensorflow-lite-image-classification-model-store-component.md) | >=2.1.0 <2.2.0 | Hard | 
| [TensorFlow Lite](tensorflow-lite-component.md) | >=2.5.0 <2.6.0 | Hard | 

------

## Configuration
<a name="tensorflow-lite-object-detection-component-configuration"></a>

This component provides the following configuration parameters that you can customize when you deploy the component.

`accessControl`  
<a name="ml-config-accesscontrol-desc"></a>(Optional) The object that contains the [authorization policy](interprocess-communication.md#ipc-authorization-policies) that allows the component to publish messages to the default notifications topic.   
Default:   

```
{
   "aws.greengrass.ipc.mqttproxy": {
      "aws.greengrass.TensorFlowLiteObjectDetection:mqttproxy:1": {
         "policyDescription": "Allows access to publish via topic ml/tflite/object-detection.",
         "operations": [
            "aws.greengrass#PublishToIoTCore"
         ],
         "resources": [
            "ml/tflite/object-detection"
         ]
      }
   }
}
```

`PublishResultsOnTopic`  
<a name="ml-config-publishresultsontopic-desc"></a>(Optional) The topic on which you want to publish the inference results. If you modify this value, then you must also modify the value of `resources` in the `accessControl` parameter to match your custom topic name.  
Default: `ml/tflite/object-detection`

`Accelerator`  <a name="ml-config-accelerator"></a>
The accelerator that you want to use. Supported values are `cpu` and `gpu`.  
The sample models in the dependent model component support only CPU acceleration. To use GPU acceleration with a different custom model, [create a custom model component](ml-customization.md#override-public-model-store) to override the public model component.  
Default: `cpu`

`ImageDirectory`  
<a name="ml-config-imagedirectory-desc"></a>(Optional) The path of the folder on the device where inference components read images. You can modify this value to any location on your device to which you have read/write access.  
<a name="ml-config-imagedirectory-obj-default"></a>Default: `/greengrass/v2/packages/artifacts-unarchived/component-name/object_detection/sample_images/`  
If you set the value of `UseCamera` to `true`, then this configuration parameter is ignored. 

`ImageName`  
<a name="ml-config-imagename-desc"></a>(Optional) The name of the image that the inference component uses as an input to a make prediction. The component looks for the image in the folder specified in `ImageDirectory`. By default, the component uses the sample image in the default image directory. AWS IoT Greengrass supports the following image formats: `jpeg`, `jpg`, `png`, and `npy`.   
<a name="ml-config-imagename-obj-default"></a>Default: `objects.jpg`  
If you set the value of `UseCamera` to `true`, then this configuration parameter is ignored. 

`InferenceInterval`  <a name="ml-config-inferenceinterval"></a>
(Optional) The time in seconds between each prediction made by the inference code. The sample inference code runs indefinitely and repeats its predictions at the specified time interval. For example, you can change this to a shorter interval if you want to use images taken by a camera for real-time prediction.  
Default: `3600`

`ModelResourceKey`  <a name="ml-config-modelresourcekey"></a>
<a name="ml-config-modelresourcekey-desc"></a>(Optional) The models that are used in the dependent public model component. Modify this parameter only if you override the public model component with a custom component.   
Default:  

```
{
    "model": "TensorFlowLite-SSD"
}
```

`UseCamera`  <a name="ml-config-usecamera"></a>
(Optional) String value that defines whether to use images from a camera connected to the Greengrass core device. Supported values are `true` and `false`.  
When you set this value to `true`, the sample inference code accesses the camera on your device and runs inference locally on the captured image. The values of the `ImageName` and `ImageDirectory` parameters are ignored. Make sure that the user running this component has read/write access to the location where the camera stores captured images.  
Default: `false`  
When you view the recipe of this component, the `UseCamera` configuration parameter doesn't appear in the default configuration. However, you can modify the value of this parameter in a [configuration merge update](update-component-configurations.md) when you deploy the component.   
When you set `UseCamera` to `true`, you must also create a symlink to enable the inference component to access your camera from the virtual environment that is created by the runtime component. For more information about using a camera with the sample inference components, see [Update component configurations](ml-tutorial-image-classification-camera.md).

**Note**  <a name="ml-config-not-visible-note"></a>
When you view the recipe of this component, the `UseCamera` configuration parameter doesn't appear in the default configuration. However, you can modify the value of this parameter in a [configuration merge update](update-component-configurations.md) when you deploy the component.   
When you set `UseCamera` to `true`, you must also create a symlink to enable the inference component to access your camera from the virtual environment that is created by the runtime component. For more information about using a camera with the sample inference components, see [Update component configurations](ml-tutorial-image-classification-camera.md).

## Local log file
<a name="tensorflow-lite-object-detection-component-log-file"></a>

This component uses the following log file.

------
#### [ Linux ]

```
/greengrass/v2/logs/aws.greengrass.TensorFlowLiteObjectDetection.log
```

------
#### [ Windows ]

```
C:\greengrass\v2\logs\aws.greengrass.TensorFlowLiteObjectDetection.log
```

------

**To view this component's logs**
+ Run the following command on the core device to view this component's log file in real time. Replace `/greengrass/v2` or *C:\$1greengrass\$1v2* with the path to the AWS IoT Greengrass root folder.

------
#### [ Linux ]

  ```
  sudo tail -f /greengrass/v2/logs/aws.greengrass.TensorFlowLiteObjectDetection.log
  ```

------
#### [ Windows (PowerShell) ]

  ```
  Get-Content C:\greengrass\v2\logs\aws.greengrass.TensorFlowLiteObjectDetection.log -Tail 10 -Wait
  ```

------

## Changelog
<a name="tensorflow-lite-object-detection-component-changelog"></a>

The following table describes the changes in each version of the component.


|  Version  |  Changes  | 
| --- | --- | 
|  2.1.12  |  Version updated for Greengrass nucleus 2.12.5 release.  | 
|  2.1.11  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  2.1.10  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  2.1.9  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  2.1.8  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  2.1.7  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  2.1.6  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  2.1.5  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  2.1.4  |  Version updated for Greengrass nucleus version 2.5.0 release.  | 
|  2.1.3  |  Version updated for Greengrass nucleus version 2.4.0 release.  | 
|  2.1.2  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  2.1.1  |  <a name="changelog-tensorflow-lite-object-detection-2.1.1"></a>[\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/tensorflow-lite-object-detection-component.html)  | 
|  2.1.0  |  Initial version.  | 

# TensorFlow Lite image classification model store
<a name="tensorflow-lite-image-classification-model-store-component"></a>

The TensorFlow Lite image classification model store (`variant.TensorFlowLite.ImageClassification.ModelStore`) is a machine learning model component that contains a pre-trained MobileNet v1 model as a Greengrass artifact. The sample model used in this component is fetched from the [TensorFlow Hub](https://tfhub.dev/) and implemented using [TensorFlow Lite](https://www.tensorflow.org/lite/guide/python).

The [TensorFlow Lite image classification](tensorflow-lite-image-classification-component.md) inference component uses this component as a dependency for the model source. To use a custom-trained TensorFlow Lite model, [create a custom version](ml-customization.md#override-public-model-store) of this model component, and include your custom model as a component artifact. You can use the recipe of this component as a template to create custom model components. 

**Topics**
+ [Versions](#tensorflow-lite-image-classification-model-store-component-versions)
+ [Type](#tensorflow-lite-image-classification-model-store-component-type)
+ [Operating system](#tensorflow-lite-image-classification-model-store-component-os-support)
+ [Requirements](#tensorflow-lite-image-classification-model-store-component-requirements)
+ [Dependencies](#tensorflow-lite-image-classification-model-store-component-dependencies)
+ [Configuration](#tensorflow-lite-image-classification-model-store-component-configuration)
+ [Local log file](#tensorflow-lite-image-classification-model-store-component-log-file)
+ [Changelog](#tensorflow-lite-image-classification-model-store-component-changelog)

## Versions
<a name="tensorflow-lite-image-classification-model-store-component-versions"></a>

This component has the following versions:
+ 2.1.x

## Type
<a name="tensorflow-lite-image-classification-model-store-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="tensorflow-lite-image-classification-model-store-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="tensorflow-lite-image-classification-model-store-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

## Dependencies
<a name="tensorflow-lite-image-classification-model-store-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#tensorflow-lite-image-classification-model-store-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 2.1.11 and 2.1.12  ]

The following table lists the dependencies for version 2.1.11 and 2.1.12 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 

------
#### [ 2.1.10 ]

The following table lists the dependencies for version 2.1.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 

------
#### [ 2.1.9 ]

The following table lists the dependencies for version 2.1.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 

------
#### [ 2.1.8 ]

The following table lists the dependencies for version 2.1.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 

------
#### [ 2.1.7 ]

The following table lists the dependencies for version 2.1.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 

------
#### [ 2.1.6 ]

The following table lists the dependencies for version 2.1.6 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 

------
#### [ 2.1.5 ]

The following table lists the dependencies for version 2.1.5 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 

------
#### [ 2.1.4 ]

The following table lists the dependencies for version 2.1.4 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 

------
#### [ 2.1.3 ]

The following table lists the dependencies for version 2.1.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 

------
#### [ 2.1.2 ]

The following table lists the dependencies for version 2.1.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 

------
#### [ 2.1.1 ]

The following table lists the dependencies for version 2.1.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 

------
#### [ 2.1.0 ]

The following table lists the dependencies for version 2.1.0 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 

------

## Configuration
<a name="tensorflow-lite-image-classification-model-store-component-configuration"></a>

This component doesn't have any configuration parameters.

## Local log file
<a name="tensorflow-lite-image-classification-model-store-component-log-file"></a>

This component doesn't output logs.

## Changelog
<a name="tensorflow-lite-image-classification-model-store-component-changelog"></a>

The following table describes the changes in each version of the component.


|  Version  |  Changes  | 
| --- | --- | 
|  2.1.12  |  Version updated for Greengrass nucleus 2.12.5 release.  | 
|  2.1.11  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  2.1.10  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  2.1.9  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  2.1.8  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  2.1.7  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  2.1.6  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  2.1.5  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  2.1.4  |  Version updated for Greengrass nucleus version 2.5.0 release.  | 
|  2.1.3  |  Version updated for Greengrass nucleus version 2.4.0 release.  | 
|  2.1.2  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  2.1.1  |  Version updated for Greengrass nucleus version 2.2.0 release.  | 
|  2.1.0  |  Initial version.  | 

# TensorFlow Lite object detection model store
<a name="tensorflow-lite-object-detection-model-store-component"></a>

The TensorFlow Lite object detection model store (`variant.TensorFlowLite.ObjectDetection.ModelStore`) is a machine learning model component that contains a pre-trained Single Shot Detection (SSD) MobileNet model as a Greengrass artifact. The sample model used in this component is fetched from the [TensorFlow Hub](https://tfhub.dev/) and implemented using [TensorFlow Lite](https://www.tensorflow.org/lite/guide/python).

The [TensorFlow Lite object detection](tensorflow-lite-object-detection-component.md) inference component uses this component as a dependency for the model source. To use a custom-trained TensorFlow Lite model, [create a custom version](ml-customization.md#override-public-model-store) of this model component, and include your custom model as a component artifact. You can use the recipe of this component as a template to create custom model components. 

**Topics**
+ [Versions](#tensorflow-lite-object-detection-model-store-component-versions)
+ [Type](#tensorflow-lite-object-detection-model-store-component-type)
+ [Operating system](#tensorflow-lite-object-detection-model-store-component-os-support)
+ [Requirements](#tensorflow-lite-object-detection-model-store-component-requirements)
+ [Dependencies](#tensorflow-lite-object-detection-model-store-component-dependencies)
+ [Configuration](#tensorflow-lite-object-detection-model-store-component-configuration)
+ [Local log file](#tensorflow-lite-object-detection-model-store-component-log-file)
+ [Changelog](#tensorflow-lite-object-detection-model-store-component-changelog)

## Versions
<a name="tensorflow-lite-object-detection-model-store-component-versions"></a>

This component has the following versions:
+ 2.1.x

## Type
<a name="tensorflow-lite-object-detection-model-store-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="tensorflow-lite-object-detection-model-store-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="tensorflow-lite-object-detection-model-store-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

## Dependencies
<a name="tensorflow-lite-object-detection-model-store-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#tensorflow-lite-object-detection-model-store-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 2.1.11 and 2.1.12 ]

The following table lists the dependencies for version 2.1.11 and 2.1.12 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 

------
#### [ 2.1.10 ]

The following table lists the dependencies for version 2.1.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 

------
#### [ 2.1.9 ]

The following table lists the dependencies for version 2.1.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 

------
#### [ 2.1.8 ]

The following table lists the dependencies for version 2.1.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 

------
#### [ 2.1.7 ]

The following table lists the dependencies for version 2.1.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 

------
#### [ 2.1.6 ]

The following table lists the dependencies for version 2.1.6 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 

------
#### [ 2.1.5 ]

The following table lists the dependencies for version 2.1.5 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 

------
#### [ 2.1.4 ]

The following table lists the dependencies for version 2.1.4 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 

------
#### [ 2.1.3 ]

The following table lists the dependencies for version 2.1.3 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 

------
#### [ 2.1.2 ]

The following table lists the dependencies for version 2.1.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 

------
#### [ 2.1.1 ]

The following table lists the dependencies for version 2.1.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 

------
#### [ 2.1.0 ]

The following table lists the dependencies for version 2.1.0 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 

------

## Configuration
<a name="tensorflow-lite-object-detection-model-store-component-configuration"></a>

This component doesn't have any configuration parameters.

## Local log file
<a name="tensorflow-lite-object-detection-model-store-component-log-file"></a>

This component doesn't output logs.

## Changelog
<a name="tensorflow-lite-object-detection-model-store-component-changelog"></a>

The following table describes the changes in each version of the component.


|  Version  |  Changes  | 
| --- | --- | 
|  2.1.12  |  Version updated for Greengrass nucleus 2.12.5 release.  | 
|  2.1.11  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  2.1.10  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  2.1.9  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  2.1.8  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  2.1.7  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  2.1.6  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  2.1.5  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  2.1.4  |  Version updated for Greengrass nucleus version 2.5.0 release.  | 
|  2.1.3  |  Version updated for Greengrass nucleus version 2.4.0 release.  | 
|  2.1.2  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  2.1.1  |  Version updated for Greengrass nucleus version 2.2.0 release.  | 
|  2.1.0  |  Initial version.  | 

# TensorFlow Lite runtime
<a name="tensorflow-lite-component"></a>

The TensorFlow Lite runtime component (`variant.TensorFlowLite`) contains a script that installs [TensorFlow Lite](https://www.tensorflow.org/lite/guide/python) version 2.5.0 and its dependencies in a virtual environment on your device. The [TensorFlow Lite image classification](tensorflow-lite-image-classification-component.md) and [TensorFlow Lite object detection](tensorflow-lite-object-detection-component.md) component use this runtime component as a dependency for installing TensorFlow Lite. 

**Note**  
TensorFlow Lite runtime component v2.5.6 and later reinstalls existing installations of the TensorFlow Lite runtime and its dependencies. This reinstallation helps to ensure that the core device runs compatible versions of TensorFlow Lite and its dependencies.

To use a different runtime, you can use the recipe of this component as a template to [create a custom machine learning component](ml-customization.md).

**Topics**
+ [Versions](#tensorflow-lite-component-versions)
+ [Type](#tensorflow-lite-component-type)
+ [Operating system](#tensorflow-lite-component-os-support)
+ [Requirements](#tensorflow-lite-component-requirements)
+ [Dependencies](#tensorflow-lite-component-dependencies)
+ [Configuration](#tensorflow-lite-component-configuration)
+ [Usage](#tensorflow-lite-component-usage)
+ [Local log file](#tensorflow-lite-component-log-file)
+ [Changelog](#tensorflow-lite-component-changelog)

## Versions
<a name="tensorflow-lite-component-versions"></a>

This component has the following versions:
+ 2.5.x

## Type
<a name="tensorflow-lite-component-type"></a>

<a name="public-component-type-generic"></a>This <a name="public-component-type-generic-phrase"></a>component is a generic component (`aws.greengrass.generic`). The [Greengrass nucleus](greengrass-nucleus-component.md) runs the component's lifecycle scripts.

<a name="public-component-type-more-information"></a>For more information, see [Component types](develop-greengrass-components.md#component-types).

## Operating system
<a name="tensorflow-lite-component-os-support"></a>

This component can be installed on core devices that run the following operating systems:
+ Linux
+ Windows

## Requirements
<a name="tensorflow-lite-component-requirements"></a>

This component has the following requirements:<a name="ml-component-requirements"></a>
+ On Greengrass core devices running Amazon Linux 2 or Ubuntu 18.04, [GNU C Library](https://www.gnu.org/software/libc/) (glibc) version 2.27 or later installed on the device.
+ On Armv7l devices, such as Raspberry Pi, dependencies for OpenCV-Python installed on the device. Run the following command to install the dependencies.

  ```
  sudo apt-get install libopenjp2-7 libilmbase23 libopenexr-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libgtk-3-0 libwebp-dev
  ```
+ Raspberry Pi devices that run Raspberry Pi OS Bullseye must meet the following requirements:
  + NumPy 1.22.4 or later installed on the device. Raspberry Pi OS Bullseye includes an earlier version of NumPy, so you can run the following command to upgrade NumPy on the device.

    ```
    pip3 install --upgrade numpy
    ```
  + The legacy camera stack enabled on the device. Raspberry Pi OS Bullseye includes a new camera stack that is enabled by default and isn't compatible, so you must enable the legacy camera stack.<a name="raspberry-pi-bullseye-enable-legacy-camera-stack"></a>

**To enable the legacy camera stack**

    1. Run the following command to open the Raspberry Pi configuration tool.

       ```
       sudo raspi-config
       ```

    1. Select **Interface Options**.

    1. Select **Legacy camera** to enable the legacy camera stack.

    1. Reboot the Raspberry Pi.

### Endpoints and ports
<a name="tensorflow-lite-component-endpoints"></a>

By default, this component uses an installer script to install packages using the `apt`, `yum`, `brew`, and `pip` commands, depending on what platform the core device uses. This component must be able to perform outbound requests to various package indexes and repositories to run the installer script. To allow this component's outbound traffic through a proxy or firewall, you must identify the endpoints for the package indexes and repositories where your core device connects to install.

Consider the following when you identify endpoints required for this component's install script:
+ The endpoints depend on the core device's platform. For example, a core device that runs Ubuntu uses `apt` rather than `yum` or `brew`. Additionally, devices that use the same package index might have different source lists, so they might retrieve packages from different repositories.
+ The endpoints might differ between multiple devices that use the same package index, because each device has its own source lists that define where to retrieve packages.
+ The endpoints might change over time. Each package index provides the URLs of the repositories where you download packages, and the owner of a package can change what URLs the package index provides.

For more information about the dependencies that this component installs, and how to disable the installer script, see the [UseInstaller](#tensorflow-lite-component-config-useinstaller-term) configuration parameter.

For more information about endpoints and ports required for basic operation, see [Allow device traffic through a proxy or firewall](allow-device-traffic.md).

## Dependencies
<a name="tensorflow-lite-component-dependencies"></a>

When you deploy a component, AWS IoT Greengrass also deploys compatible versions of its dependencies. This means that you must meet the requirements for the component and all of its dependencies to successfully deploy the component. This section lists the dependencies for the [released versions](#tensorflow-lite-component-changelog) of this component and the semantic version constraints that define the component versions for each dependency. You can also view the dependencies for each version of the component in the [AWS IoT Greengrass console](https://console.aws.amazon.com//greengrass). On the component details page, look for the **Dependencies** list.

------
#### [ 2.5.14 and 2.5.15 ]

The following table lists the dependencies for version 2.5.14 and 2.5.15 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.13.0 | Soft | 

------
#### [ 2.5.13 ]

The following table lists the dependencies for version 2.5.13 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.12.0 | Soft | 

------
#### [ 2.5.12 ]

The following table lists the dependencies for version 2.5.12 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.11.0 | Soft | 

------
#### [ 2.5.11 ]

The following table lists the dependencies for version 2.5.11 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.10.0 | Soft | 

------
#### [ 2.5.10 ]

The following table lists the dependencies for version 2.5.10 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.9.0 | Soft | 

------
#### [ 2.5.9 ]

The following table lists the dependencies for version 2.5.9 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.8.0 | Soft | 

------
#### [ 2.5.8 ]

The following table lists the dependencies for version 2.5.8 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.7.0 | Soft | 

------
#### [ 2.5.5 - 2.5.7 ]

The following table lists the dependencies for versions 2.5.5 through 2.5.7 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.6.0 | Soft | 

------
#### [ 2.5.3 and 2.5.4 ]

The following table lists the dependencies for versions 2.5.3 and 2.5.4 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.5.0 | Soft | 

------
#### [ 2.5.2 ]

The following table lists the dependencies for version 2.5.2 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.4.0 | Soft | 

------
#### [ 2.5.1 ]

The following table lists the dependencies for version 2.5.1 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.3.0 | Soft | 

------
#### [ 2.5.0 ]

The following table lists the dependencies for version 2.5.0 of this component.


| Dependency | Compatible versions | Dependency type | 
| --- | --- | --- | 
| [Greengrass nucleus](greengrass-nucleus-component.md) | >=2.0.0 <2.2.0 | Soft | 

------

For more information about component dependencies, see the [component recipe reference](component-recipe-reference.md#recipe-reference-component-dependencies).

## Configuration
<a name="tensorflow-lite-component-configuration"></a>

This component provides the following configuration parameters that you can customize when you deploy the component.

`MLRootPath`  
<a name="ml-config-mlrootpath-desc"></a>(Optional) The path of the folder on Linux core devices where inference components read images and write inference results. You can modify this value to any location on your device to which the user running this component has read/write access.  
<a name="ml-config-mlrootpath-default-tfl"></a>Default: `/greengrass/v2/work/variant.TensorFlowLite/greengrass_ml`

`WindowsMLRootPath`  
This feature is available in v1.6.6 and later of this component.  
<a name="ml-config-windowsmlrootpath-desc"></a>(Optional) The path of the folder on Windows core device where inference components read images and write inference results. You can modify this value to any location on your device to which the user running this component has read/write access.  
<a name="ml-config-windowsmlrootpath-default-dlr"></a>Default: `C:\greengrass\v2\\work\\variant.DLR\\greengrass_ml`

  `UseInstaller`   
<a name="ml-config-useinstaller-desc-tfl"></a>(Optional) String value that defines whether to use the installer script in this component to install TensorFlow Lite and its dependencies. Supported values are `true` and `false`.   <a name="ml-config-useinstaller-libraries-tfl"></a>

Set this value to `false` if you want to use a custom script for TensorFlow Lite installation, or if you want to include runtime dependencies in a pre-built Linux image. To use this component with the AWS-provided TensorFlow Lite inference components, install the following libraries, including any dependencies, and make them available to the system user, such as `ggc_user`, that runs the ML components.
+ [Python](https://www.python.org/downloads/) 3.8 or later, including `pip` for your version of Python
+ [TensorFlow Lite](https://www.tensorflow.org/lite/guide/python) v2.5.0
+ [NumPy](https://numpy.org/install/)
+ [OpenCV-Python](https://pypi.org/project/opencv-python/)
+ [AWS IoT Device SDK v2 for Python](https://github.com/aws/aws-iot-device-sdk-python-v2)
+ [AWS Common Runtime (CRT) Python](https://github.com/awslabs/aws-crt-python)
+ [Picamera](https://picamera.readthedocs.io/en/release-1.13/) (for Raspberry Pi devices)
+ [`awscam` module](https://docs.aws.amazon.com/deeplens/latest/dg/deeplens-library-awscam-module.html) (for AWS DeepLens devices)
+ libGL (for Linux devices)
<a name="ml-config-useinstaller-default"></a>Default: `true`

## Usage
<a name="tensorflow-lite-component-usage"></a>

Use this component with the `UseInstaller` configuration parameter set to `true` to install TensorFlow Lite and its dependencies on your device. The component sets up a virtual environment on your device that includes the OpenCV and NumPy libraries that are required for TensorFlow Lite. 

**Note**  <a name="ml-installer-component-usage-note"></a>
The installer script in this component also installs the latest versions of additional system libraries that are required to configure the virtual environment on your device and to use the installed machine learning framework. This might upgrade the existing system libraries on your device. Review the following table for the list of libraries that this component installs for each supported operating system. If you want to customize this installation process, set the `UseInstaller` configuration parameter to `false`, and develop your own installer script.


| Platform | Libraries installed on the device system | Libraries installed in the virtual environment | 
| --- | --- | --- | 
| Armv7l | build-essential, cmake, ca-certificates, git | setuptools, wheel | 
| Amazon Linux 2 | mesa-libGL | None | 
| Ubuntu | wget | None | 

When you deploy your inference component, this runtime component first verifies if your device already has TensorFlow Lite and its dependencies installed. If not, then the runtime component installs them for you. 

## Local log file
<a name="tensorflow-lite-component-log-file"></a>

This component uses the following log file.

------
#### [ Linux ]

```
/greengrass/v2/logs/variant.TensorFlowLite.log
```

------
#### [ Windows ]

```
C:\greengrass\v2\logs\variant.TensorFlowLite.log
```

------

**To view this component's logs**
+ Run the following command on the core device to view this component's log file in real time. Replace `/greengrass/v2` or *C:\$1greengrass\$1v2* with the path to the AWS IoT Greengrass root folder.

------
#### [ Linux ]

  ```
  sudo tail -f /greengrass/v2/logs/variant.TensorFlowLite.log
  ```

------
#### [ Windows (PowerShell) ]

  ```
  Get-Content C:\greengrass\v2\logs\variant.TensorFlowLite.log -Tail 10 -Wait
  ```

------

## Changelog
<a name="tensorflow-lite-component-changelog"></a>

The following table describes the changes in each version of the component.


|  Version  |  Changes  | 
| --- | --- | 
|  2.5.15  |  Version updated for Greengrass nucleus 2.12.5 release.  | 
|  2.5.14  | Version updated for Greengrass nucleus version 2.12.0 release. | 
|  2.5.13  | Version updated for Greengrass nucleus version 2.11.0 release. | 
|  2.5.12  | Version updated for Greengrass nucleus version 2.10.0 release. | 
|  2.5.11  | Version updated for Greengrass nucleus version 2.9.0 release. | 
|  2.5.10  | Version updated for Greengrass nucleus version 2.8.0 release. | 
|  2.5.9  |  Version updated for Greengrass nucleus version 2.7.0 release.  | 
|  2.5.8  |  Version updated for Greengrass nucleus version 2.6.0 release.  | 
|  2.5.7  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/tensorflow-lite-component.html)  | 
|  2.5.6  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/tensorflow-lite-component.html)  | 
|  2.5.5  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/tensorflow-lite-component.html)  | 
|  2.5.4  |  [\[See the AWS documentation website for more details\]](http://docs.aws.amazon.com/greengrass/v2/developerguide/tensorflow-lite-component.html)  | 
|  2.5.3  |  Version updated for Greengrass nucleus version 2.4.0 release.  | 
|  2.5.2  |  Version updated for Greengrass nucleus version 2.3.0 release.  | 
|  2.5.1  |  Version updated for Greengrass nucleus version 2.2.0 release.  | 
|  2.5.0  |  Initial version.  |