Plot fails with `NoneType object not callable` for plotly backend in pandas

Thanks for your interest in Plotly.py!

Before opening an issue, please search for existing and closed issues :)

Please accompany bug reports with a reproducible example. Please use the latest version of plotly.py in your report unless not applicable.

When plotting a DataFrame with pandas and the plotly backend this fails only in debug mode with the following error:

Traceback (most recent call last):
  File "/Users/patrick/mambaforge/envs/random/lib/python3.10/site-packages/numpy/core/getlimits.py", line 650, in __init__
    self.dtype = numeric.dtype(int_type)
TypeError: 'NoneType' object is not callable

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "<frozen importlib._bootstrap>", line 1027, in _find_and_load
  File "<frozen importlib._bootstrap>", line 1006, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 688, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 883, in exec_module
  File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed
  File "/Users/patrick/mambaforge/envs/random/lib/python3.10/site-packages/plotly/express/__init__.py", line 15, in <module>
    from ._imshow import imshow
  File "/Users/patrick/mambaforge/envs/random/lib/python3.10/site-packages/plotly/express/_imshow.py", line 4, in <module>
    from .imshow_utils import rescale_intensity, _integer_ranges, _integer_types
  File "/Users/patrick/mambaforge/envs/random/lib/python3.10/site-packages/plotly/express/imshow_utils.py", line 21, in <module>
    _integer_ranges = {t: (np.iinfo(t).min, np.iinfo(t).max) for t in _integer_types}
  File "/Users/patrick/mambaforge/envs/random/lib/python3.10/site-packages/plotly/express/imshow_utils.py", line 21, in <dictcomp>
    _integer_ranges = {t: (np.iinfo(t).min, np.iinfo(t).max) for t in _integer_types}
  File "/Users/patrick/mambaforge/envs/random/lib/python3.10/site-packages/numpy/core/getlimits.py", line 652, in __init__
    self.dtype = numeric.dtype(type(int_type))
TypeError: 'NoneType' object is not callable

# import plotly.express as px
import pandas as pd

pd.options.plotting.backend = "plotly"

df = pd.DataFrame({"a": [1, 2, 3], "b": 100})
fig = df.plot()
fig.write_html(Path("/tmp").joinpath("testplot.html"))

I am using pandas 1.4.3, numpy 1.23.1 and saw this on poorly 5.10 and back to 5.7 (stopped checking then). I suspect that this is similar to https://github.com/numpy/numpy/issues/21008.

The error disappears then importing plotly.express before calling df.plot(). (Importing only plotly is not sufficient).

Could not find anything similar on the issue tracker

Note that GitHub Issues are meant to be used for bug reports and feature requests only. Implementation or usage questions should be asked on community.plotly.com or on Stack Overflow (tagged plotly).

Edit: Sorry for the issue title, auto spelling...

Beginne mit der minimalen pandas-DataFrame-Reproduktion unter Verwendung des plotly-Plotting-Backends und vergleiche sie damit, zuerst plotly.express zu importieren. Untersuche den Traceback-Pfad in plotly/express/_imshow.py, plotly/express/imshow_utils.py und numpy/core/getlimits.py. Als erledigt gilt die Aufgabe, wenn df.plot() im Debug-Modus keinen NoneType-Fehler mehr auslöst, ohne dass zuvor ein plotly.express-Import erforderlich ist.