ave() averages the values of all argument rows into a single trace.

ave() averages the values of all argument rows into a single trace.

count() counts the non-missing values of all argument rows.

fill() fills in all the missing datapoints with nearby points.

filter() returns a filtered set of Rows that match the given query. It expects a single argument that is a string in URL query format, such as: os=Ubuntu12&config=8888.

geo() folds the values of all argument rows into a single geometric mean trace.

log() applies a base-10 logarithm to the datapoints.

norm() normalizes the rows to a mean of 0 and a standard deviation of 1.0. If a second optional number is passed in to norm() then that is used as the minimum standard deviation that is normalized, otherwise it defaults to 0.1.

ratio(a, b) returns the point by point ratio of two rows. That is, it returns a trace with a[i]/b[i] for every point in a and b.

Computes a new trace that is scaled by 1/(ave) where ave is the average of the input trace.

shortcut() returns a set of Rows that match the given shortcut id. It expects a single argument that is the id of a shortcut set of traces.

Computes the step function, i.e the ratio of the ave of the first half of the trace divided by the ave of the second half of the trace.

Sum() Sums the values of all argument rows into a single trace.

Computes the mean for all the values in a trace and return a trace where every value is that mean.

Computes the mean for all the values in a trace and return a trace where every value is that mean.

Computes the Coefficient of Variation for all the values in a trace and return a trace where every value is that CoV.

Computes the std dev for all the values in a trace and return a trace where every value is that stddev.

`ave(fill(filter("config=8888")))`

Plot a single curve that's the average of all the traces that have a config of '8888'. The traces have missing data points filled in from nearby runs.

`ave(filter(""))`

Plot a single curve that's the average of all the traces.

`norm(filter("test=desk_linkedin.skp_1_1000_1000"))`

Plot the normalized version of all the traces for 'desk_linkedin.skp_1_1000_1000'.