# Allora API: How to Query Data of Existing Topics

The **Allora API** provides an interface to query real-time on-chain data of the latest inferences made by workers. Here's an explanation of how it works using the example endpoint:

**Generic**: `https://allora-api.testnet.allora.network/emissions/{version_number}/latest_network_inferences/{topic_id}`

**Example**: `https://allora-api.testnet.allora.network/emissions/v4/latest_network_inferences/1`

Where:

- "v4" represents the latest network version number
- "1" represents the topic ID

Sample Response:

```
{
"network_inferences": {
"topic_id": "1",
"reputer_request_nonce": null,
"reputer": "",
"extra_data": null,
"combined_value": "2605.533879185080648394998043723508",
"inferer_values": [
{
"worker": "allo102ksu3kx57w0mrhkg37kvymmk2lgxqcan6u7yn",
"value": "2611.01109296"
},
{
"worker": "allo10q6hm2yae8slpvvgmxqrcasa30gu5qfysp4wkz",
"value": "2661.505295679922"
}
],
"forecaster_values": [
{
"worker": "allo1za8r9v0st4ntfyeka23qs5wvd7mvsnzhztupk0",
"value": "2610.160000000000000000000000000000"
}
],
"naive_value": "2605.533879185080648394998043723508",
"one_out_inferer_values": [
{
"worker": "allo102ksu3kx57w0mrhkg37kvymmk2lgxqcan6u7yn",
"value": "2570.859434973857748387096774193548"
},
{
"worker": "allo10q6hm2yae8slpvvgmxqrcasa30gu5qfysp4wkz",
"value": "2569.230589724828006451612903225806"
}
],
"one_out_forecaster_values": [],
"one_in_forecaster_values": [],
"one_out_inferer_forecaster_values": []
},
"inferer_weights": [
{
"worker": "allo102ksu3kx57w0mrhkg37kvymmk2lgxqcan6u7yn",
"weight": "0.0002191899319465528034563075461505151"
},
{
"worker": "allo10q6hm2yae8slpvvgmxqrcasa30gu5qfysp4wkz",
"weight": "0.0002191899319465528034563075461505151"
}
],
"forecaster_weights": [
{
"worker": "allo1za8r9v0st4ntfyeka23qs5wvd7mvsnzhztupk0",
"weight": "0.1444137067859501612197657742201029"
}
],
"forecast_implied_inferences": [
{
"worker": "allo1za8r9v0st4ntfyeka23qs5wvd7mvsnzhztupk0",
"value": "2610.160000000000000000000000000000"
}
],
"inference_block_height": "1349577",
"loss_block_height": "0",
"confidence_interval_raw_percentiles": [
"2.28",
"15.87",
"50",
"84.13",
"97.72"
],
"confidence_interval_values": [
"2492.1675618299669694181830608795809",
"2543.9249467952655499150756965734158",
"2611.033130351115229549044053766836",
"2662.29523395638446190095015123294396",
"2682.827040221238"
]
}
```

Please be aware that there may be some expected volatility in predictions due to the nascency of the network and the more forgiving testnet configurations currently in place. We are actively working on implementing an outlier protection mechanism, which will be applied at the consumer layer and tailored to individual use cases in the near future.

## Breaking Down the Response

Below is an explanation of important sub-objects displayed in the JSON output:

`topic_id`

In this case, "1" represents the topic being queried. Topics define the context and rules for a particular inference.

`naive_value`

The **naive value** omits all forecast-implied inferences from the weighted average by setting their weights to zero. The naive network inference is used to quantify the contribution of the
forecasting task to the network accuracy, which in turn sets the reward distribution between the inference and forecasting tasks.

`combined_value`

The **combined value** is an optimized inference that represents a collective intelligence approach, taking both naive submissions and forecast data into account.

If you are looking to just get one value or number from Allora for a data oracle, this would be the one to take.

`inferer_values`

Workers in the network submit their inferences, each represented by an `allo`

address. For example:

```
{
"worker": "allo102ksu3kx57w0mrhkg37kvymmk2lgxqcan6u7yn",
"value": "2611.01109296"
}
```

Each worker submits a value based on their own models. These individual submissions contribute to both the naive and combined values. The combined value gives higher weighting to more reliable workers, based on performance or other criteria.

`one_out_inferer_values`

These values simulate removing a single worker from the pool to see how the overall inference changes. This is a technique used to evaluate the impact of individual inferences on the combined result.

`forecast_implied_inferences`

The Forecast-Implied Inference uses forecasted losses and worker inferences to produce a predicted value where each prediction is weighted based on how accurately the forecasters predicted losses in previous time steps, or epochs.

`inference_block_height`

The specific chain block that the inference data was generated

`confidence_interval_raw_percentiles`

Fixed percentiles that are used to generate confidence intervals

`confidence_interval_values`

Confidence intervals show the predicted range of outcomes based on worker inferences.