Seaborn histplot with one element gives "`bins` must be positive" error

[AnavSood]

For certain values of x and y > 0, if I run sns.histplot([x], binwidth=y) I get ValueError: `bins` must be positive, when an integer. For example, run sns.histplot([0.49928], binwidth=0.03).

See here for discussion: https://stackoverflow.com/questions/70553730/seaborn-histplot-with-one-element-gives-bins-must-be-positive-error

[mwaskom]

Hi, your link doesn't go anywhere. Please provide a reproducible example in the issue, thanks.

[AnavSood]

Hi, your link doesn't go anywhere. Please provide a reproducible example in the issue, thanks.

Apologies! Updated on both accounts.

[mwaskom]

Stacktrace:

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Input In [106], in <module>
----> 1 sns.histplot([0.49928], binwidth=0.03)

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/seaborn/distributions.py:1462, in histplot(data, x, y, hue, weights, stat, bins, binwidth, binrange, discrete, cumulative, common_bins, common_norm, multiple, element, fill, shrink, kde, kde_kws, line_kws, thresh, pthresh, pmax, cbar, cbar_ax, cbar_kws, palette, hue_order, hue_norm, color, log_scale, legend, ax, **kwargs)
   1451 estimate_kws = dict(
   1452     stat=stat,
   1453     bins=bins,
   (...)
   1457     cumulative=cumulative,
   1458 )
   1460 if p.univariate:
-> 1462     p.plot_univariate_histogram(
   1463         multiple=multiple,
   1464         element=element,
   1465         fill=fill,
   1466         shrink=shrink,
   1467         common_norm=common_norm,
   1468         common_bins=common_bins,
   1469         kde=kde,
   1470         kde_kws=kde_kws,
   1471         color=color,
   1472         legend=legend,
   1473         estimate_kws=estimate_kws,
   1474         line_kws=line_kws,
   1475         **kwargs,
   1476     )
   1478 else:
   1480     p.plot_bivariate_histogram(
   1481         common_bins=common_bins,
   1482         common_norm=common_norm,
   (...)
   1492         **kwargs,
   1493     )

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/seaborn/distributions.py:441, in _DistributionPlotter.plot_univariate_histogram(self, multiple, element, fill, common_norm, common_bins, shrink, kde, kde_kws, color, legend, line_kws, estimate_kws, **plot_kws)
    438     weights = None
    440 # Do the histogram computation
--> 441 heights, edges = estimator(observations, weights=weights)
    443 # Rescale the smoothed curve to match the histogram
    444 if kde and key in densities:

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/seaborn/_statistics.py:386, in Histogram.__call__(self, x1, x2, weights)
    384 """Count the occurrences in each bin, maybe normalize."""
    385 if x2 is None:
--> 386     return self._eval_univariate(x1, weights)
    387 else:
    388     return self._eval_bivariate(x1, x2, weights)

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/seaborn/_statistics.py:364, in Histogram._eval_univariate(self, x, weights)
    361     bin_kws = self.define_bin_params(x, weights=weights, cache=False)
    363 density = self.stat == "density"
--> 364 hist, bin_edges = np.histogram(
    365     x, **bin_kws, weights=weights, density=density,
    366 )
    368 if self.stat == "probability" or self.stat == "proportion":
    369     hist = hist.astype(float) / hist.sum()

File <__array_function__ internals>:180, in histogram(*args, **kwargs)

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/numpy/lib/histograms.py:793, in histogram(a, bins, range, normed, weights, density)
    681 r"""
    682 Compute the histogram of a dataset.
    683 
   (...)
    789 
    790 """
    791 a, weights = _ravel_and_check_weights(a, weights)
--> 793 bin_edges, uniform_bins = _get_bin_edges(a, bins, range, weights)
    795 # Histogram is an integer or a float array depending on the weights.
    796 if weights is None:

File ~/miniconda3/envs/py39/lib/python3.9/site-packages/numpy/lib/histograms.py:424, in _get_bin_edges(a, bins, range, weights)
    421         raise TypeError(
    422             '`bins` must be an integer, a string, or an array') from e
    423     if n_equal_bins < 1:
--> 424         raise ValueError('`bins` must be positive, when an integer')
    426     first_edge, last_edge = _get_outer_edges(a, range)
    428 elif np.ndim(bins) == 1:

ValueError: `bins` must be positive, when an integer

seaborn version: '0.11.2'
numpy version: '1.22.0'

Relevant seaborn code:

https://github.com/mwaskom/seaborn/blob/v0.11.2/seaborn/_statistics.py#L256-L258

[vmodi1]

Hey @AnavSood, I am new to github and would like to make a contribution to this issue.

[sajozsattila]

I have the same problem. Example code:

sns.histplot([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
                     stat="probability", binwidth=0.05)

[cmayer]

#2785 and #2721 have in common that data point(s) are not counted even though they are there.
In the one case this produces an error message, in the other case an output that omits points.