Mathematical Operations

.abs() Return the absolute value of each value in the series list
.add() Adds the values of one or more series in a seriesList to each position, in each series, of the input seriesList
.cusum() Return the cumulative sum of a series, starting at a base.
.derivative() Plot the change in values over time.
.divide() Divides the values of one or more series in a seriesList to each position, in each series, of the input seriesList
.elasticsearch() Pull data from an elasticsearch instance
.es() Pull data from an elasticsearch instance
.first() This is an internal function that simply returns the input seriesList. Don't use this
.fit() Fills null values using a defined fit function
.hide() Hide the series by default
.holt() Sample the beginning of a series and use it to forecast what should happen via several optional parameters. In general, like everything, this is crappy at predicting the future. You're much better off using it to predict what should be happening right now, for the purpose of anomaly detection. Note that nulls will be filled with forecasted values. Deal with it.
.lines() Show the seriesList as lines
.log() Return the logarithm value of each value in the series list (default base: 10)
.max() Maximum values of one or more series in a seriesList to each position, in each series, of the input seriesList
.min() Minimum values of one or more series in a seriesList to each position, in each series, of the input seriesList
.movingaverage() Calculate the moving average over a given window. Nice for smoothing noisey series
.movingstd() Calculate the moving standard deviation over a given window. Uses naive two-pass algorithm. Rounding errors may become more noticeable with very long series, or series with very large numbers.
.multiply() Multiply the values of one or more series in a seriesList to each position, in each series, of the input seriesList
.mvavg() Calculate the moving average over a given window. Nice for smoothing noisey series
.mvstd() Calculate the moving standard deviation over a given window. Uses naive two-pass algorithm. Rounding errors may become more noticeable with very long series, or series with very large numbers.
.plus() Adds the values of one or more series in a seriesList to each position, in each series, of the input seriesList
.points() Show the series as points
.precision() number of digits to round the decimal portion of the value to
.range() Changes the max and min of a series while keeping the same shape
.scale_interval() Changes scales a value (usually a sum or a count) to a new interval. For example, as a per-second rate
.static() Draws a single value across the chart
.subtract() Subtract the values of one or more series in a seriesList to each position, in each series, of the input seriesList
.sum() Adds the values of one or more series in a seriesList to each position, in each series, of the input seriesList
.trend() Draws a trend line using a specified regression algorithm
.trim() Set N buckets at the start or end of a series to null to fit the “partial bucket issue”
.value() Draws a single value across the chart

Logical Operations

.condition() Compares each point to a number, or the same point in another series using an operator, then sets its valueto the result if the condition proves true, with an optional else.
.if()

.condition() and .if() can perform the same function

takes 4 arguments:

  • operator One of lt, lte, eq, gt or gte.
  • “if” You’ll be comparing each point in the original series to this value using the operator you specified.
  • “then” Set the point to this value if the condition matches
  • “else” set the point to a value if the condition does not match.

If point is less than 500 then nullify it

.es().if(lt, 500, null)

Graph Operations

.bars() Show the seriesList as bars
.color() Change the color of the series
.label() Change the label of the series. Use %s reference the existing label
.legend() Set the position and style of the legend on the plot
.title() Adds a title to the top of the plot. If called on more than 1 seriesList the last call will be used.
.yaxis() Configures a variety of y-axis options, the most important likely being the ability to add an Nth (eg 2nd) y-axis

Misc

.props() Use at your own risk, sets arbitrary properties on the series. For example .props(label=bears!)
.graphite() [experimental] Pull data from graphite. Configure your graphite server in Kibana's Advanced Settings
.quandl() [experimental] Pull data from quandl.com using the quandl code. Set “timelion:quandl.key” to your free API key in Kibana's Advanced Settings. The API has a really low rate limit without a key.
.wb() [experimental] Pull data from http://data.worldbank.org/ using path to series. The worldbank provides mostly yearly data, and often has no data for the current year. Try offset=-1y if you get no data for recent time ranges.
.wbi() [experimental] Pull data from http://data.worldbank.org/ using the country name and indicator. The worldbank provides mostly yearly data, and often has no data for the current year. Try offset=-1y if you get no data for recent time ranges.
.worldbank() [experimental] Pull data from http://data.worldbank.org/ using path to series. The worldbank provides mostly yearly data, and often has no data for the current year. Try offset=-1y if you get no data for recent time ranges.
 
timelion.txt · Last modified: 2017/10/06 08:19 by root
 
RSS - 200 © CrosswireDigitialMedia Ltd