On October 7, 2026, AWS will discontinue support for Amazon Lookout for Equipment. After October 7, 2026, you will no longer be able to access the Lookout for Equipment console or resources. For more information, [see the following](https://aws.amazon.com/blogs/machine-learning/preserve-access-and-explore-alternatives-for-amazon-lookout-for-equipment/).
# Scheduling inference
**Note**
You can also schedule inference [with the AWS SDK for Python (Boto)](SDK-examples.md#create-model-sdk).
## Starting inference
After you create a model, you can use it to monitor your asset in real time. To use your model to monitor your asset, you do the following.
To schedule inference, you specify the model, the schedule, the Amazon S3 location of where the model is reading the data, and where it outputs the results of the inference.
1. Sign in to AWS Management Console and open the Amazon Lookout for Equipment console at [Amazon Lookout for Equipment console](https://console.aws.amazon.com/lookoutequipment/home).
1. Choose **Models**. Then choose the model that monitors your asset.
1. Choose **Schedule inference**.
1. For **Inference schedule name**, specify the name for the inference schedule.
1. For **Model**, choose the model that is monitoring the data coming from your asset.
1. For **S3 location** under **Input data**, specify the Amazon S3 location of the input data coming from the asset.
1. For **Data upload frequency**, specify how often your asset sends the data to the Amazon S3 bucket.
1. For **S3 location** under **Output data**, specify the Amazon S3 location to store the output of the inference results.
1. For **IAM role** under **Access Permissions**, specify the IAM role that provides Amazon Lookout for Equipment with access to your data in Amazon S3.
1. Choose **Schedule inference**.
# Managing inference schedules
## Stopping inference
This section explains how to halt the inference process.
1. From the AWS console, under Lookout for Equipment, from the left nav, choose **Inference schedules**.
1. If necessary, choose the **Active schedules** tab.
1. Select the schedule that you want to stop.
1. Choose **Stop**.
1. Choose **Stop schedule**.
1. Your stopped schedule will appear on the **Inactive schedules** tab.
## Resuming inference
This section explains how to resume a stopped inference schedule.
1. From the AWS console, under Lookout for Equipment, from the left nav, choose **Inference schedules**.
1. If necessary, choose the **Inactive schedules** tab.
1. Choose **Set as active**.
1. Your stopped schedule will appear on the **Active schedules** tab.
## Editing an active schedule
1. From the AWS console, under Lookout for Equipment, from the left nav, choose **Inference schedules**.
1. If necessary, choose the **Active schedules** tab.
1. Select the schedule that you want to edit.
1. Choose **Edit**.
1. On the pop-up window, choose **edit**.
**Note**
After you finish editing an inference schedule, the schedule returns to the activation status that it was in before you started editing.
A schedule that was active before editing will return to active status after editing.
## Editing an inactive schedule
1. From the AWS console, under Lookout for Equipment, from the left nav, choose **Inference schedules**.
1. If necessary, choose the **Inactive schedules** tab.
1. Select the schedule that you want to edit.
1. Choose **Edit**.
1. On the pop-up window, choose **edit**.
**Note**
After you finish editing an inference schedule, the schedule returns to the activation status that it was in before you started editing.
A schedule that was inactive before editing will remain inactive after editing. To re-activate it, you must select the schedule on the **Inactive schedules** page and choose **Set as active**.
## Delete an active schedule
1. From the AWS console, under Lookout for Equipment, from the left nav, choose **Inference schedules**.
1. If necessary, choose the **Active schedules** tab.
1. Select the schedule that you want to delete.
1. Choose **Delete**.
1. In the pop-up window, choose **Stop** to indicate that you are going to stop the schedule before deleting it.
1. In the pop-up window, enter *delete* in the text field.
1. In the pop-up window, choose **delete**.
## Delete an inactive schedule
1. From the AWS console, under Lookout for Equipment, from the left nav, choose **Inference schedules**.
1. If necessary, choose the **Inactive schedules** tab.
1. Select the schedule that you want to delete.
1. Choose **Delete**.
1. In the pop-up window, enter *delete* in the text field.
1. In the pop-up window, choose **delete**.
# Understanding the inference process
When you're planning your use of Lookout for Equipment, it may be useful to understand exactly what happens at each step of the inference process.
## Understanding inference scheduling windows
+ When you schedule inference, you may set your data upload frequency time to any of the following values, in minutes: 5, 10, 15, 30, 60.
+ Lookout for Equipment then calculates the base number of segments per hour by dividing 60 by the length of your segments.
+ You may also set an offset window in increments of minutes, from 0 to 60.
+ At the beginning of each segment, Lookout for Equipment waits for the offset window to close before running inference.
+ At the top of the hour, the process begins again.
| Inference interval | Inferences per hour | First inference after 09:00 (with no offset) | First inference after 09:00 (with a 5-minute offset) |
| --- | --- | --- | --- |
| 5 | 12 | 09:05 | 09:10 |
| 10 | 6 | 09:10 | 09:15 |
| 15 | 4 | 09:15 | 09:20 |
| 30 | 2 | 09:30 | 09:35 |
| 60 | 1 | 10:00 | 10:05 |
## The inference process
![\[Inference steps\]](http://docs.aws.amazon.com/lookout-for-equipment/latest/ug/images/inference-steps.png)
1. Lookout for Equipment looks for the component name (which can be the name of an asset or a sensor, depending on how your data was ingested).
1. Once the component name is found in the file name, Lookout for Equipment looks at the time stamp in the CSV file name.
1. The timestamp in the file name must be within the range of time that your scheduler is running. For example, if the scheduler is running every 5 minutes, then at 9:05, Lookout for Equipment will look for any files that have a timestamp from 9:00 to 9:05. Any files with timestamps outside this range will be ignored for the inference run.
1. Lookout for Equipment automatically ingests the files with the right component name, and within the right time range.
1. Lookout for Equipment opens the CSV file and runs inference on any rows in the CSV file with timestamps that fit within the scheduler window. For example, if the scheduler is running every 5 minutes, and the current time is 9:05, then Lookout for Equipment will grab any files with the timestamp in the file name from 9:00 to 9:05, and will then run inference on any rows in the CSV with timestamps between 9:00 to 9:05.
1. The inference results are placed into your designated output bucket in a JSON file.
1. The steps above are repeated in perpetuity until the scheduler is turned off.