search

thinger-io / thinger-server

Thinger.io Server Repository
MIT License
2 stars 0 forks source link

Option to Filter bucket data when viewing it on the Dashboard #88

Closed georgevbsantiago closed 2 months ago

georgevbsantiago commented 6 months ago

The bucket data filter option to view in the widget (time series, tachometer, text...) would be important to omit noisy measurements. The intention is that the user places the Filter parameters in the Widget and noisy measurements are not visualized in the widget (time series, tachometer, text...). The Filter option is different from the "Transform" option. If there are noisy measurements, the "Transform" option will be influenced. The Filter option aims to exclude noisy measurements from the graph display and improve the "Transform" calculation.

It may be possible to implement the Filter in Influxdb, as stated in this part of the documentation: https://docs.influxdata.com/influxdb/v2/query-data/influxql/functions/technical-analysis/#technical-analysis-functions

KAUFMANS_ADAPTIVE_MOVING_AVERAGE()

Kaufman’s Adaptive Moving Average (KAMA) is a moving average designed to account for sample noise or volatility. KAMA will closely follow data points when the data swings are relatively small and noise is low. KAMA will adjust when the data swings widen and follow data from a greater distance. This trend-following indicator can be used to identify the overall trend, time turning points and filter data movements. Source

image image

georgevbsantiago commented 6 months ago

Hi @alvarolb

Is it feasible to implement a Filter option with Influxdb so that noisy data is not visualized in the widget graph when the source is the bucket? It would be important for the functionality to be made available for the Time Series and for KPIs such as the Tachometer

georgevbsantiago commented 6 months ago

We need an approach (a filter option) that allows us to use algorithms and parameters that eliminate noisy measurements from the widget visualization (time series, tachometer...). The Aggregation and Transformation option does not work for us, as noisy readings impact the value results.

image

georgevbsantiago commented 6 months ago

Just a supplementary note.

A "noisy measurement" is not always bad. In the case of distance sensors, a "noisy measurement" may indicate that there is an object close to the sensor that needs to be moved away from the sensor. A "noisy measurement" can sometimes signal the need for some inversion in the context close to the sensor. However, for most customers, they do not need to see this information on the chart. Seeing "noisy measurements" is of most interest to technicians operating the equipment. For the client (e.g. decision-making manager) he wants to see the graph without the "noisy measurements". The technician, when viewing noisy measurements in other dashboards or analysis, will know that it is necessary to adjust or repair the sensor that generated the "noisy measurement".

georgevbsantiago commented 2 months ago

Hi @alvarolb

The "Dashboard Functions" functionality in version 6.0 implemented this functionality, correct? Can we close?