<!-- {"title": "Import JSON", "category": "time-series", "complexity": "basic", "order": "3"} -->
|
|
<!-- readme
|
### Notes
|
|
* JSON is a native format for Label Studio and it allows to store time series data right in LS task.
|
<a target="_blank" href="https://labelstud.io/guide/tasks.html">Read more about LS tasks in documentation.</a>
|
|
* JSON is necessary when you need to combine multiple data source in one task.
|
For example, you mix two CSV files for time series into one labeling task or
|
if you try to build Text + TimeSeries configuration.
|
|
### Quickstart
|
|
1. Prepare JSON-formatted input:
|
|
```json
|
{
|
"ts": {
|
"time": [0,1,2],
|
"first_column": [3,4,5],
|
"second_column": [6,7,8]
|
}
|
}
|
```
|
|
2. Set your data keys in the following attributes:
|
* `timeColumn="time"`
|
* `column="first_column"` and `column="second_column"`
|
|
3. Add more `<Channel>` tags for more columns if needed
|
|
Read more about
|
<a href="https://labelstud.io/tags/timeseries.html" target="_blank">TimeSeries</a>,
|
<a href="https://labelstud.io/tags/timeserieslabels.html" target="_blank">TimeSeriesLabels</a>,
|
<a href="https://labelstud.io/tags/timeseries.html#Channel" target="_blank">Channel</a>.
|
-->
|
|
<View>
|
<Header value="Time Series from JSON"
|
style="font-weight: normal"/>
|
|
<TimeSeriesLabels name="label" toName="ts">
|
<Label value="Anomaly" background="#a4a"/>
|
<Label value="Ordinary" background="#aa4"/>
|
</TimeSeriesLabels>
|
|
<TimeSeries timeColumn="time"
|
name="ts" value="$ts" valueType="json">
|
<Channel column="first_column"
|
strokeColor="#1f77b4"/>
|
<Channel column="second_column"
|
strokeColor="#ff7f0e"/>
|
</TimeSeries>
|
</View>
|
