plot with float_contrast=False causes pandas groupby error

[dcnadler]

Using jupyter notebook in jupyter/scipy-notebook docker container
DockerFile

FROM jupyter/scipy-notebook:2ce7c06a61a1
RUN pip install dabest

Using example from github:

import pandas as pd
import dabest

iris = pd.read_csv("https://github.com/mwaskom/seaborn-data/raw/master/iris.csv")
iris_dabest = dabest.load(data=iris, x="species", y="petal_width",
                          idx=("setosa", "versicolor", "virginica"))
iris_dabest.mean_diff.plot()

output:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-4-eb33a02ff120> in <module>
      1 # Produce a Cumming estimation plot.
----> 2 iris_dabest.mean_diff.plot()

/opt/conda/lib/python3.7/site-packages/dabest/_classes.py in plot(self, color_col, raw_marker_size, es_marker_size, swarm_label, contrast_label, swarm_ylim, contrast_ylim, custom_palette, swarm_desat, halfviolin_desat, halfviolin_alpha, float_contrast, show_pairs, group_summaries, group_summaries_offset, fig_size, dpi, swarmplot_kwargs, violinplot_kwargs, slopegraph_kwargs, reflines_kwargs, group_summary_kwargs, legend_kwargs)
   1233         del all_kwargs["self"]
   1234 
-> 1235         out = EffectSizeDataFramePlotter(self, **all_kwargs)
   1236 
   1237         return out

/opt/conda/lib/python3.7/site-packages/dabest/plotter.py in EffectSizeDataFramePlotter(EffectSizeDataFrame, **plot_kwargs)
    375                          gap_width_percent=1.5,
    376                          type=group_summaries, ax=rawdata_axes,
--> 377                          **group_summary_kwargs)
    378 
    379 

/opt/conda/lib/python3.7/site-packages/dabest/plot_tools.py in gapped_lines(data, x, y, type, offset, ax, line_color, gap_width_percent, **kwargs)
    160 
    161     medians   = data.groupby(x)[y].median().reindex(index=group_order)
--> 162     quantiles = data.groupby(x)[y].quantile([0.25, 0.75])\
    163                                   .unstack()\
    164                                   .reindex(index=group_order)

/opt/conda/lib/python3.7/site-packages/pandas/core/groupby/groupby.py in quantile(self, q, interpolation)
   1906             post_processing=post_processor,
   1907             q=q,
-> 1908             interpolation=interpolation,
   1909         )
   1910 

/opt/conda/lib/python3.7/site-packages/pandas/core/groupby/groupby.py in _get_cythonized_result(self, how, grouper, aggregate, cython_dtype, needs_values, needs_mask, needs_ngroups, result_is_index, pre_processing, post_processing, **kwargs)
   2246                 func = partial(func, ngroups)
   2247 
-> 2248             func(**kwargs)  # Call func to modify indexer values in place
   2249 
   2250             if result_is_index:

pandas/_libs/groupby.pyx in pandas._libs.groupby.group_quantile()

TypeError: must be real number, not list

I have noticed this error tends to happen when using float_contrast=False (either specified with two samples, or when using more than two samples).

[josesho]

Hi @dcnadler,

This seems suspiciously familiar to #52. Are you using pandas==0.25.0?

[dcnadler]

@josesho, yes pandas 0.25.0

[josesho]

Can you try either downgrading to 0.24 or upgrading to 0.25.1?