Chart types¶

pymyIO renders 34 chart types — the same set as R-myIO. Every type below is a valid value for MyIO.add_layer(type=...).

Basic¶

Type	Required mapping	Notes
`line`	`x_var`, `y_var`	Add `group` to stack per-series lines.
`point`	`x_var`, `y_var`	Pair with `transform="lm"` for a fitted line.
`bar`	`x_var`, `y_var`	Call `.define_categorical_axis(x_axis=True)` for categorical x.
`groupedBar`	`x_var`, `y_var`, `group`	One bar per group, dodged.
`area`	`x_var`, `low_y`, `high_y`	Shaded band between `low_y` and `high_y`.
`histogram`	`value`	Bin counts.

Statistical¶

Type	Required mapping	Notes
`regression`	`x_var`, `y_var`	Composite: points + fit line. `options={"method": "lm" \\| "polynomial"}`.
`rangeBar`	`x_var`, `low_y`, `high_y`	:material-alert: `transform="mean_ci"` doesn't work yet — see Roadmap.
`qq`	`y_var`	Composite: reference line + points (envelope deferred). Pass `mapping["group"]` for per-group Q-Q.
`boxplot`	`x_var`, `y_var`	Composite: IQR rangeBar + whisker points + median + Tukey outliers.
`violin`	`x_var`, `y_var`	Composite: per-group KDE area + optional box / median / raw points.
`ridgeline`	`x_var`, `y_var`, `group`	Composite: stacked density areas with `overlap` factor.
`comparison`	`x_var`, `y_var`	:material-alert: uses `pairwise_test` — see Roadmap.

Specialized¶

Type	Required mapping	Notes
`donut`	`x_var`, `y_var`	Annular pie.
`gauge`	`value`	Single-value dial.
`treemap`	`level_1`, `level_2`	Hierarchical rectangles.
`heatmap`	`x_var`, `y_var`, `value`	Color-encoded matrix.
`hexbin`	`x_var`, `y_var`, `radius`	2D density binning.
`calendarHeatmap`	`date`, `value`	GitHub-style contribution calendar.

Financial¶

Type	Required mapping	Notes
`candlestick`	`x_var`, `open`, `high`, `low`, `close`	OHLC bars.
`waterfall`	`x_var`, `y_var`	Bridge chart; `total` column marks terminal bars.

Relational¶

Type	Required mapping	Notes
`sankey`	`source`, `target`, `value`	Flow diagram.
`parallel`	`dimensions`	Parallel-coordinates plot.

"New" family¶

Type	Required mapping	Notes
`lollipop`	`x_var`, `y_var`	Bar+dot hybrid.
`dumbbell`	`x_var`, `low_y`, `high_y`	Two-point range per category.
`waffle`	`category`, `value`	10×10 square grid by share.
`beeswarm`	`x_var`, `y_var`	Jittered points with force-directed nudging.
`bump`	`x_var`, `y_var`, `group`	Rank transitions over time.
`radar`	`axis`, `value`	Polar plot across named axes.
`funnel`	`stage`, `value`	Sales / pipeline funnel.

Deferred / not-yet-rendering in 0.1.x¶

See Roadmap for landing targets.

Type	Reason
`comparison`	Needs `pairwise_test` transform port.
`survfit`	Needs Kaplan-Meier composite expansion (transform exists, expansion not wired).
`histogram_fit`	Needs `composite_histogram_fit` port (transform exists, expansion not wired).

Composite types¶

regression, boxplot, violin, ridgeline, comparison, qq, survfit, and histogram_fit expand to multiple primitive layers under the hood (e.g. regression = point + line[lm]). You do not need to stack them manually — MyIO.add_layer(type="regression", …) handles it.

Layer compatibility¶

myIO groups layer types by axis style. Compatible groups can coexist on the same chart; incompatible ones raise at add_layer() time:

axes-continuous: line, point, area, candlestick, rangeBar, text, regression, bracket, comparison, qq, bump, survfit
axes-categorical: bar, groupedBar, waterfall, boxplot, violin, lollipop, dumbbell, beeswarm
axes-binned: histogram, ridgeline, histogram_fit
axes-matrix: heatmap, calendarHeatmap
axes-hex: hexbin
standalone-flow: sankey, treemap, waffle, funnel, parallel
standalone-radial: donut, gauge, radar