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/).
# Reviewing inference results
After you've scheduled inference, you are able to see how your equipment is operating.
**Topics**
+ [Reviewing inference results in the console](understanding-results-console.md)
+ [Reviewing inference results in a JSON file](understanding-results-json.md)
# Reviewing inference results in the console
## Using the main inference schedules page
On the inference schedules main page you'll find your list of inference schedules, both active and inactive (on different tabs). For each schedule, you'll find the model name, data upload frequency, and latest results.
In this context, *latest results* means the results from the most recent inference run.
![\[Inference schedules table showing active schedules for various equipment models with status indicators.\]](http://docs.aws.amazon.com/lookout-for-equipment/latest/ug/images/inference-schedules-main.png)
To edit, delete, stop, or restart a schedule, see [Managing inference schedules](managing-inference-schedules.md).
## Using the inference schedule detail page
On the inference schedule detail page you'll find details about the anomalous behavior of your assets, as presented in the context of a particular inference schedule.
You'll also find metadata about the schedule itself.
![\[Inference schedule overview for pump8-inference showing status, model, and 7-day results.\]](http://docs.aws.amazon.com/lookout-for-equipment/latest/ug/images/inference-schedule-detail.png)
At the top of the results tab are the 7-day inference results. These results provide information about anomalous behavior that occurred over the past week.
*Latest results* refers to results from the latest inference run.
*7-day results* indicates the percentage of hours during the last seven days, during which an anomaly was detected.
Use the slider to zoom in on a particular event (red bar).
Click on a particular event (red bar) to view details about it.
After you click on a particular event, the **Event details** tab indicates which sensors contributed the most to that event.
![\[Bar chart showing top 10 contributing sensors, with Temperature1 at 27.5% and Vibration1 at 16.7%.\]](http://docs.aws.amazon.com/lookout-for-equipment/latest/ug/images/inference-event-details.png)
**Note**
Lookout for Equipment only records events that last longer than 5 minutes.
# Reviewing inference results in a JSON file
The JSON file containing the inference results is stored in the Amazon Simple Storage Service (Amazon S3) bucket that you've specified.
For the sensor data that your asset sends to Amazon S3, Amazon Lookout for Equipment marks the group of readings as either normal or abnormal. For each group of abnormal readings, you can see the sensors that Lookout for Equipment used to indicate that the equipment is behaving abnormally.
The following shows example JSON output.
```
{"timestamp": "2021-03-11T22:24:00.000000", "prediction": 0, "prediction_reason": "MACHINE_OFF"}
{"timestamp": "2021-03-11T22:25:00.000000", "prediction": 1, "prediction_reason": "ANOMALY_DETECTED", "anomaly_score": 0.72385, "diagnostics": [{"name": "component_5feceb66\\sensor0", "value": 0.02346}, {"name": "component_5feceb66\\sensor1", "value": 0.10011}, {"name": "component_5feceb66\\sensor2", "value": 0.11162}, {"name": "component_5feceb66\\sensor3", "value": 0.14419}, {"name": "component_5feceb66\\sensor4", "value": 0.12219}, {"name": "component_5feceb66\\sensor5", "value": 0.14936}, {"name": "component_5feceb66\\sensor6", "value": 0.17829}, {"name": "component_5feceb66\\sensor7", "value": 0.00194}, {"name": "component_5feceb66\\sensor8", "value": 0.05446}, {"name": "component_5feceb66\\sensor9", "value": 0.11437}]}
{"timestamp": "2021-03-11T22:26:00.000000", "prediction": 0, "prediction_reason": "NO_ANOMALY_DETECTED", "anomaly_score": 0.41227, "diagnostics": [{"name": "component_5feceb66\\sensor0", "value": 0.03533}, {"name": "component_5feceb66\\sensor1", "value": 0.24063}, {"name": "component_5feceb66\\sensor2", "value": 0.06327}, {"name": "component_5feceb66\\sensor3", "value": 0.08303}, {"name": "component_5feceb66\\sensor4", "value": 0.18598}, {"name": "component_5feceb66\\sensor5", "value": 0.10839}, {"name": "component_5feceb66\\sensor6", "value": 0.08721}, {"name": "component_5feceb66\\sensor7", "value": 0.06792}, {"name": "component_5feceb66\\sensor8", "value": 0.1309}, {"name": "component_5feceb66\\sensor9", "value": 0.07735}]}
{"timestamp": "2021-03-11T22:27:00.000000", "prediction": 0, "prediction_reason": "NO_ANOMALY_DETECTED", "anomaly_score": 0.10541, "diagnostics": [{"name": "component_5feceb66\\sensor0", "value": 0.02533}, {"name": "component_5feceb66\\sensor1", "value": 0.34063}, {"name": "component_5feceb66\\sensor2", "value": 0.07327}, {"name": "component_5feceb66\\sensor3", "value": 0.03303}, {"name": "component_5feceb66\\sensor4", "value": 0.18598}, {"name": "component_5feceb66\\sensor5", "value": 0.10839}, {"name": "component_5feceb66\\sensor6", "value": 0.08721}, {"name": "component_5feceb66\\sensor7", "value": 0.06792}, {"name": "component_5feceb66\\sensor8", "value": 0.1309}, {"name": "component_5feceb66\\sensor9", "value": 0.07735}]}
{"timestamp": "2021-03-11T22:28:00.000000", "prediction": 0, "prediction_reason": "NO_ANOMALY_DETECTED", "anomaly_score": 0.24867, "diagnostics": [{"name": "component_5feceb66\\sensor0", "value": 0.04533}, {"name": "component_5feceb66\\sensor1", "value": 0.14063}, {"name": "component_5feceb66\\sensor2", "value": 0.08327}, {"name": "component_5feceb66\\sensor3", "value": 0.07303}, {"name": "component_5feceb66\\sensor4", "value": 0.18598}, {"name": "component_5feceb66\\sensor5", "value": 0.10839}, {"name": "component_5feceb66\\sensor6", "value": 0.08721}, {"name": "component_5feceb66\\sensor7", "value": 0.06792}, {"name": "component_5feceb66\\sensor8", "value": 0.1309}, {"name": "component_5feceb66\\sensor9", "value": 0.07735}]}
{"timestamp": "2021-03-11T22:29:00.000000", "prediction": 1, "prediction_reason": "ANOMALY_DETECTED", "anomaly_score": 0.79376, "diagnostics": [{"name": "component_5feceb66\\sensor0", "value": 0.04533}, {"name": "component_5feceb66\\sensor1", "value": 0.14063}, {"name": "component_5feceb66\\sensor2", "value": 0.08327}, {"name": "component_5feceb66\\sensor3", "value": 0.07303}, {"name": "component_5feceb66\\sensor4", "value": 0.18598}, {"name": "component_5feceb66\\sensor5", "value": 0.10839}, {"name": "component_5feceb66\\sensor6", "value": 0.08721}, {"name": "component_5feceb66\\sensor7", "value": 0.06792}, {"name": "component_5feceb66\\sensor8", "value": 0.1309}, {"name": "component_5feceb66\\sensor9", "value": 0.07735}]}
```
For the `prediction` field, a `value` of `1` indicates abnormal equipment behavior. A `value` of `0` indicates normal equipment behavior.
If the value of `prediction_reason` isn't `MACHINE_OFF`, Amazon Lookout for Equipment returns an object that contains a diagnostics list, regardless of the value of `prediction`. The `diagnostics` list has the name of the sensors and the weights of the sensors' contributions in indicating abnormal equipment behavior. For each sensor, the `name` field indicates the name of the sensor. The `value` field indicates the percentage of the sensor's contribution to the prediction value. By seeing the percentage of each sensor's contribution to the prediction value, you can see how the data from each sensor was weighted.
The anomaly score is a value between 0 and 1 that indicates the intensity of the anomaly.
The prediction reason can be ANOMALY\$1DETECTED, NO\$1ANOMALY\$1DETECTED or MACHINE\$1OFF.