myforestplot package

Submodules

myforestplot.cleaning_utils module

myforestplot.cleaning_utils.add_pretty_risk_column(res, risk, lower, upper, fml='.2f', ref='Ref.')[source]

Add prrety risk string column.

Parameters

  • res (DataFrame) – Dataframe contaning points and confidence intervals.

  • risk (str) – point estimates of risk column name.

  • lower (str) – lower confidence interval column name.

  • upper (str) – upper confidence interval column name.

  • fml (str) – formula for f string.

  • ref (str) – if point esitmate column is empty, insert this string.

Return type

Series

myforestplot.cleaning_utils.count_category_frequency(df, categorical_cols, impute_continuous=True)[source]

Count category frequency.

Parameters

  • df (DataFrame) – Original dataframe.

  • categorical_cols (List[str]) – Columns for categorical variables.

  • impute_continuous (bool) – columns not specified as categorical_cols were imputed for item and number of observations (nobs).

Return type

DataFrame

myforestplot.cleaning_utils.sort_category_item(df_, order, item_order=None)[source]

Sort category and item based on categorical values.

Parameters

  • df – dataframe containing category and item.

  • categorical – Dictionary containing column names and its order of items.

  • order (List[str]) – if specified, category is ordered based on this variable.

Return type

DataFrame

myforestplot.cleaning_utils.statsmodels_fitting_result_dataframe(res, alpha=0.05, accessor=<ufunc 'exp'>)[source]

Create category and item columns from the statsmodels result. Categorical results are divided into original column name (category) and its items (item).

Parameters

  • res – statsmodels’ fitting results.

  • alpha (float) – The significance level for the confidence interval.

  • accessor (Callable[[array], array]) – Function to access each model result, which is summarized and displayed.

Return type

DataFrame

myforestplot.cleaning_utils.statsmodels_pretty_result_dataframe(data, res, order, cont_cols=None, item_order=None, fml='.2f', accessor=<ufunc 'exp'>)[source]

Obtain pretty result dataframe from statsmodels results. Fitting coefficients are converted by np.exp.

Parameters

  • data (DataFrame) – original dataframe.

  • res – statsmodels results.

  • categorical – Dictionary containing column names and its order of items.

  • order (List[str]) – if specified, category is ordered based on this variable.

  • fml (str) – formula for f string of pretty risk.

  • accessor (Callable[[array], array]) – Function to access each model result, which is summarized and displayed.

Return type

DataFrame

myforestplot.myforestplot module

class myforestplot.myforestplot.ForestPlot(df, ratio=(8, 3), fig_ax_index=None, figsize=(5, 3), yticks_show=False, yticklabels_show=False, xticks_show=True, text_axis_off=True, hide_spines=<factory>, dpi=150, vertical_align=False)[source]

Bases: object

Parameters

  • df (DataFrame) – Result dataframe.

  • ratio (Tuple[float, float]) – Number of axes field and size ratio of these axes. self.axd contains axes of which index starts from 1.

  • fig_ax_index (Optional[List[int]]) – If specified, x ticks and x labels are left.

  • figsize (Tuple[float, float]) – Figure size.

  • hide_spines (List[str]) – Hide outlines of axes. Takes “right”,”top”,”bottom”,”left” or these list.

  • vertical_align (bool) – Align categorical names above items. It requires dataframe to have “category” and “item” column names.

ax_method_to_figs(method, *args, **kwds)[source]

Apply axis method to all the figure fields.

Parameters

  • method (str) – Name of axis method.

  • fig_ax_index – If not specified, apply to all figs. If specified, apply to specified index axis fields.

  • *args – Passed to a specified method.

  • **kwds – Passed to a specified method.

create_y_index()[source]
df: DataFrame
dpi: int = 150
draw_horizontal_line(y, scale=0.1, **kwds)[source]

Draw horizontal line.

Parameters

kwds – Passed to ax.axhline.

draw_outer_marker(index, lower=0, upper=1, lower_marker=4, upper_marker=5, y_adj=0, df=None, log_scale=False, scale=0, **kwds)[source]

Draw markers to indicate outer range of confidence intervals.

Parameters

  • y_adj (float) – For this value, points are moved vertically.

  • scale (float) – Control position of markers. scale * x range is slided towards inside.

  • kwds – Passed to ax.scatter.

embed_cate_strings(index, col, x, header='', fontsize=None, y_header=1.0, y_adj=0.0, text_kwds=None, header_kwds=None, replace=None, df=None)[source]

Embed category values on vertically aligned positions. The position of strings become different only if self.vertical_align == True.

embed_strings(index, col, x, header='', fontsize=None, y_header=1.0, y_adj=0.0, text_kwds=None, header_kwds=None, duplicate_hide=False, replace=None, df=None)[source]

Embed strings/values of one column with header.

Parameters

  • col (str) – Column name for text.

  • x (float) – x axis value of text position, ranging from 0 to 1.

  • df (Optional[DataFrame]) – Dataframe for another result.

errorbar(index, risk='risk', lower=0, upper=1, y_adj=0, errorbar_kwds=None, ref_kwds=None, df=None, errorbar_color=None, ref_color=None, label=None, log_scale=False)[source]
Parameters

  • index (int) – axis’s index.

  • risk (str) – Column name for risk.

  • lower (Union[str, int]) – Column name for lower confidence interval.

  • upper (Union[str, int]) – Column name for upper confidence interval.

  • y_adj (float) – For this value, points are moved vertically.

  • errorbar_kwds (Optional[dict]) – Passed to ax.errorbar function.

  • ref_kwds (Optional[dict]) – Passed to ax.scatter function.

  • df (Optional[DataFrame]) – Dataframe for another result.

  • errorbar_color (Optional[str]) – If specified, ecolor and coloer in erorrbar_kwds is changed to this value.

  • ref_color (Optional[str]) – If specified, ecolor and coloer in ref_kwds is changed to this value.

  • label (Optional[str]) – Label for stratified drawings. Passed to ax.errorbar.

  • log_scale (bool) – Plot risk in log scale (np.log).

fig_ax_index: Optional[List[int]] = None
figsize: Tuple[float, float] = (5, 3)
figure_layout()[source]
h_multi_embed_strings(fig_ax_index, col, x, df, by, order, y_adj=0.0, multi_kwds=None, **kwds)[source]

Embed strings for multiple text fileds for stratified dataframe. fig_ax_index nad x takes one value or list types.

Parameters

  • fig_ax_index (Union[int, List[int]]) – Should be same as stratified group items.

  • x (Union[float, List[float]]) – x axis value of text position, ranging from 0 to 1.

  • df (DataFrame) – DataFrame to be plotted.

  • by (str) – Dataframe is stratified by this column.

  • order (List[str]) – Column items are plotted by this order.

  • col (str) – Columns to be used for strings.

  • **kwds – Passed to myforestplot.vis_utils.embed_strings_forestplot.

See also

myforestplot.vis_utils.embed_strings_forestplot myforestplot.myforestplot.ForestPlot.embed_strings myforestplot.myforestplot.ForestPlot.embed_cate_strings

h_multi_errorbar(df, by, order, y_adj=0.0, multi_kwds=None, **kwds)[source]

Horizontal multiple errorbar plots.

Parameters

  • df (DataFrame) – Dataframe to be stratified.

  • by (str) – Dataframe is stratified by this column.

  • order (List[str]) – Column items are plotted by this order.

  • y_adj (float) – For this value, points are moved vertically.

  • multi_kwds (Optional[Dict[str, list]]) – Options changed over each plotting are specified by this parameter.

  • kwds – Passsed to ForestPlot.errorbar.

hide_spines: List[str]
horizontal_variable_separators(scale=0.1, **kwds)[source]

Draw horizontal lines for seprating variables.

Parameters

kwds – Passed to ax.axhline function.

ratio: Tuple[float, float] = (8, 3)
text_axis_off: bool = True
v_multi_embed_strings(index, col, x, df, by, order, scale=0.4, multi_kwds=None, **kwds)[source]

Embed strings for multiple text fileds for stratified dataframe. fig_ax_index nad x takes one value or list types.

v_multi_errorbar(index, df, by, order, scale=0.4, multi_kwds=None, **kwds)[source]

Verticle multiple errorbar plots.

Parameters

  • index (int) – To draw points to this fig_ax_index field.

  • df (DataFrame) – Dataframe to be stratified.

  • by (str) – Dataframe is stratified by this column.

  • order (List[str]) – If specified, column items are plotted by this order.

  • scale (float) – [-scale, scale] is set to be a range of y_adj.

  • multi_kwds (Optional[Dict[str, list]]) – Options changed over each plotting are specified by this parameter.

  • kwds – Passsed to ForestPlot.errorbar.

vertical_align: bool = False
xticks_show: bool = True
yticklabels_show: bool = False
yticks_show: bool = False
class myforestplot.myforestplot.SimpleForestPlot(df, ratio=(8, 3), fig_ax_index=None, figsize=(5, 3), yticks_show=False, yticklabels_show=False, xticks_show=True, text_axis_off=True, hide_spines=<factory>, dpi=150, vertical_align=False)[source]

Bases: ForestPlot

Simple version of a forest plot, contaning one text field and one axis field.

df: DataFrame
draw_outer_marker(*args, **kwds)[source]

Draw markers to indicate outer range of confidence intervals.

Parameters

  • y_adj – For this value, points are moved vertically.

  • scale – Control position of markers. scale * x range is slided towards inside.

  • kwds – Passed to ax.scatter.

embed_cate_strings(*args, **kwds)[source]

Embed category values on vertically aligned positions. The position of strings become different only if self.vertical_align == True.

embed_strings(*args, **kwds)[source]

Embed strings/values of one column with header.

Parameters

  • col – Column name for text.

  • x – x axis value of text position, ranging from 0 to 1.

  • df – Dataframe for another result.

errorbar(*args, **kwds)[source]
Parameters

  • index – axis’s index.

  • risk – Column name for risk.

  • lower – Column name for lower confidence interval.

  • upper – Column name for upper confidence interval.

  • y_adj – For this value, points are moved vertically.

  • errorbar_kwds – Passed to ax.errorbar function.

  • ref_kwds – Passed to ax.scatter function.

  • df – Dataframe for another result.

  • errorbar_color – If specified, ecolor and coloer in erorrbar_kwds is changed to this value.

  • ref_color – If specified, ecolor and coloer in ref_kwds is changed to this value.

  • label – Label for stratified drawings. Passed to ax.errorbar.

  • log_scale – Plot risk in log scale (np.log).

hide_spines: List[str]
v_multi_embed_strings(*args, **kwds)[source]

Embed strings for multiple text fileds for stratified dataframe. fig_ax_index nad x takes one value or list types.

v_multi_errorbar(*args, **kwds)[source]

Verticle multiple errorbar plots.

Parameters

  • index – To draw points to this fig_ax_index field.

  • df – Dataframe to be stratified.

  • by – Dataframe is stratified by this column.

  • order – If specified, column items are plotted by this order.

  • scale – [-scale, scale] is set to be a range of y_adj.

  • multi_kwds – Options changed over each plotting are specified by this parameter.

  • kwds – Passsed to ForestPlot.errorbar.

myforestplot.vis_utils module

myforestplot.vis_utils.embed_strings_forestplot(ax, ser, y_index, x, header='', fontsize=None, y_header=1.0, text_kwds=None, header_kwds=None, replace=None)[source]

Embed strings/values of one column with header.

Parameters

  • ser (Series) – Values of this series will be embedded.

  • x (float) – x axis value of text position, ranging from 0 to 1.

  • df – Dataframe for another result.

myforestplot.vis_utils.errorbar_forestplot(ax, y_index, df=None, risk='risk', lower=0, upper=1, y_adj=0, errorbar_kwds=None, ref_kwds=None, errorbar_color=None, ref_color=None, label=None, log_scale=False)[source]

Error bar plot for a forest plot.

Parameters

  • ax (Axes) – Axis to be drawn.

  • y_index (array) – index to be plotted.

  • risk (str) – Column name for risk.

  • lower (Union[str, int]) – Column name for lower confidence interval.

  • upper (Union[str, int]) – Column name for upper confidence interval.

  • y_adj (float) – For this value, plotting is moved.

  • errorbar_kwds (Optional[dict]) – Passed to ax.errorbar function.

  • ref_kwds (Optional[dict]) – Passed to ax.scatter function.

  • df (Optional[DataFrame]) – Dataframe for another result.

  • label (Optional[str]) – Label for stratified drawings. Passed to ax.errorbar.

  • log_scale (bool) – Plot risk in log scale (np.log).

myforestplot.vis_utils.get_multiple_y_adjs(n, scale)[source]

For multiple vertical plotting, automatic adjustments of y_adj for y_index is needed.

Parameters

  • n (int) – Number of stratificaitons to be plotted.

  • scale (float) – [-scale, scale] is set to be a range of y_adj.

Return type

array

myforestplot.vis_utils.obtain_indexes_from_category_item(ser_cate, ser_item)[source]

Create index for category and item from series of category and item for vertically aligned labels and errorbar plot. It is noted that index has negative continuous values, starting from 0 to -n.

Parameters

  • ser_cate (Series) – Series of categories matched with ser_item.

  • ser_item (Series) – Series of items.

Return type

Tuple[array, array]

Returns

Indexes for caategories and items.

myforestplot.vis_utils.set_default_keywords(kwds, def_kwds)[source]

Set default keywords arguments.

Return type

dict

Module contents