# cytoflow.utility.algorithms¶

Useful algorithms.

`ci` – determine a confidence interval by boostrapping.

`percentiles` – find percentiles in an array.

`bootstrap` – resample (with replacement) and store aggregate values.

cytoflow.utility.algorithms.ci(data, func, which=95, boots=1000)[source]

Determine the confidence interval of a function applied to a data set by bootstrapping.

Parameters
• data (pandas.DataFrame) – The data to resample.

• func (callable) – A function that is called on a resampled `data`

• which (int) – The percentile to use for the confidence interval

• boots (int (default = 1000):) – How many times to bootstrap

Returns

The confidence interval.

Return type

(float, float)

cytoflow.utility.algorithms.percentiles(a, pcts, axis=None)[source]

Like `scipy.stats.scoreatpercentile` but can take and return array of percentiles.

Parameters
• a (array) – data

• pcts (sequence of percentile values) – percentile or percentiles to find score at

• axis (int or None) – if not None, computes scores over this axis

Returns

scores – array of scores at requested percentiles first dimension is length of object passed to `pcts`

Return type

array

cytoflow.utility.algorithms.bootstrap(*args, **kwargs)[source]

Resample one or more arrays with replacement and store aggregate values. Positional arguments are a sequence of arrays to bootstrap along the first axis and pass to a summary function.

Parameters
• n_boot (int, default 10000) – Number of iterations

• axis (int, default None) – Will pass axis to `func` as a keyword argument.

• units (array, default None) – Array of sampling unit IDs. When used the bootstrap resamples units and then observations within units instead of individual datapoints.

• smooth (bool, default False) – If True, performs a smoothed bootstrap (draws samples from a kernel destiny estimate); only works for one-dimensional inputs and cannot be used `units` is present.

• func (callable, default np.mean) – Function to call on the args that are passed in.

• random_seed (int | None, default None) – Seed for the random number generator; useful if you want reproducible resamples.

Returns

• array – array of bootstrapped statistic values