| 1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465 |
- def boxplot_single_factor(df='dataframe', column_names=None, grid=False, ar=(0, 0), axtickfontsize=9,
- axtickfontname='Arial', dim=(6, 4), show=False, figtype='png', figname='boxplot', r=300,
- ylm=None, box_line_style='-', box_line_width=1, box_line_color='b', med_line_style='-',
- med_line_width=1, med_line_color='g', whisk_line_color='b', cap_color='b',
- add_sign_symbol=False, symb_dist=None, sign_symbol_opts={'symbol': '*', 'fontsize': 8 },
- pv=None, notch=False, outliers=True, fill_box_color=True, dotplot=False, dotsize=6,
- colordot=['#101820ff'], valphadot=1, markerdot='o', theme=None):
- if theme == 'dark':
- general.dark_bg()
- plt.subplots()
- if column_names:
- xbar = column_names
- else:
- xbar = list(df.columns)
- # rot is x axis rotation
- other_args = {'grid': grid, 'rot': ar[0], 'fontsize': axtickfontsize, 'notch':notch, 'showfliers':outliers,
- 'figsize': dim, 'patch_artist': fill_box_color}
- color_args = {'medians': med_line_color, 'boxes': box_line_color, 'whiskers': whisk_line_color,
- 'caps': cap_color}
- medianprops_args = {'linestyle': med_line_style, 'linewidth': med_line_width}
- boxprops_args = {'linestyle': box_line_style, 'linewidth': box_line_width}
- if isinstance(column_names, list):
- df.boxplot(column=column_names, **other_args, boxprops=boxprops_args, medianprops=medianprops_args,
- color=color_args)
- else:
- df.boxplot(**other_args, boxprops=boxprops_args, color=color_args, medianprops=medianprops_args)
- # ylm must be tuple of start, end, interval
- if ylm:
- plt.ylim(bottom=ylm[0], top=ylm[1])
- plt.yticks(np.arange(ylm[0], ylm[1], ylm[2]), fontsize=axtickfontsize, fontname=axtickfontname)
- plt.yticks(fontsize=axtickfontsize, rotation=ar[1], fontname=axtickfontname)
- color_list_dot = colordot
- if len(color_list_dot) == 1:
- color_list_dot = colordot * len(xbar)
- # checked for unstacked data
- if dotplot:
- for cols in range(len(xbar)):
- plt.scatter(
- x=np.linspace(xbar[cols] - bw / 2, xbar[cols] + bw / 2, int(bar_counts[cols])),
- y=df[df.columns[cols]].dropna(), s=dotsize, color=color_list_dot[cols], zorder=10, alpha=valphadot,
- marker=markerdot)
- size_factor_to_start_line = max(df.max()) / 20
- if add_sign_symbol:
- # p and symb_dist should be dict
- if isinstance(pv, dict):
- for k, v in pv.items():
- if isinstance(symb_dist, dict):
- if k not in symb_dist:
- symb_dist[k] = 0
- y_pos = df[k].max() + size_factor_to_start_line + symb_dist[k]
- else:
- y_pos = df[k].max() + size_factor_to_start_line
- if y_pos > 0 and v <= 0.05:
- pv_symb = general.pvalue_symbol(v, sign_symbol_opts['symbol'])
- if pv_symb:
- plt.annotate(pv_symb, xy=((xbar.index(k))+1, y_pos),
- fontsize=sign_symbol_opts['fontsize'],
- ha="center")
- general.get_figure(show, r, figtype, figname, theme)
|
