pegasus.elbowplot¶
- pegasus.elbowplot(data, rep='pca', pval='0.05', panel_size=(6, 4), return_fig=False, dpi=300.0, **kwargs)[source]¶
Generate Elbowplot and suggest n_comps to select based on random matrix theory (see utils.largest_variance_from_random_matrix).
- Parameters
data (
UnimodalData
orMultimodalData
object.) – The main data object.rep (
str
, optional, default:pca
) – Representation to consider, either “pca” or “tsvd”.pval (
str
, optional (default: “0.05”).) – P value cutoff on the null distribution (random matrix), choosing from “0.01” and “0.05”.top_n (
int
, optional, default:20
) – Only show top_n up/down regulated pathways.panel_size (tuple, optional (default: (6, 4))) – The plot size (width, height) in inches.
return_fig (
bool
, optional, default:False
) – Return aFigure
object ifTrue
; returnNone
otherwise.dpi (
float
, optional, default:300.0
) – The resolution in dots per inch.
- Return type
Optional
[Figure
]- Returns
Figure object – A
matplotlib.figure.Figure
object containing the dot plot ifreturn_fig == True
.Update
data.uns
–{rep}_ncomps
: Recommended components to pick.
Examples
>>> fig = pg.elbowplot(data, dpi = 500)