Seaborn 0.12

[trevismd]

The newest version of Seaborn is not officially supported.

This is the place to discuss the need for its support, and potentially identified problems and/or solutions for it.

[phildong]

This is very much needed! The two things my advisor constantly ask me to do is to plot standard error and statistical stars. The former is added in seaborn 0.12. It's so sad I can't do both at the same time.

[noahsprent]

Not sure if this helps anyone else, including @phildong, but by juggling around dependencies I managed to get something that allows me to use 'errorbar' for barplots as well as statannotations:

statannotations==0.6.0
pandas==1.5.3
seaborn==0.12.0

[sjg2203]

@noahsprent how did you get that to work, I tried but still get me this error detailed here #133
Any chance anyone could help solve that one?
Cheers

[noahsprent]

Hmm, I've just tried it in a fresh conda environment with the first example from statsmodels, adding errorbar="se", and it worked. Code I ran:
`%matplotlib inline
import matplotlib.pyplot as plt
import seaborn as sns
from statannotations.Annotator import Annotator

sns.set(style="whitegrid")

df = sns.load_dataset("tips")

x = "day"
y = "total_bill"
order = ["Sun", "Thur", "Fri", "Sat"]
ax = sns.barplot(data=df, x=x, y=y, order=order, errorbar="ci")
annot = Annotator(ax, [("Thur", "Fri"), ("Thur", "Sat"), ("Fri", "Sun")], data=df, plot='violinplot', x=x, y=y, order=order)
annot.configure(test="Mann-Whitney", text_format="star", loc="outside", verbose=2)
annot.apply_test()
ax, test_results = annot.annotate()
plt.savefig("example_non-hue_outside.png", dpi=300, bbox_inches="tight")`

Output of pip freeze:
asttokens==2.4.0
backcall==0.2.0
comm==0.1.4
contourpy==1.1.1
cycler==0.12.0
debugpy==1.8.0
decorator==5.1.1
executing==2.0.0
fonttools==4.43.0
ipykernel==6.25.2
ipython==8.16.1
jedi==0.19.1
jupyter_client==8.3.1
jupyter_core==5.3.2
kiwisolver==1.4.5
matplotlib==3.8.0
matplotlib-inline==0.1.6
nest-asyncio==1.5.8
numpy==1.26.0
packaging==23.2
pandas==1.5.3
parso==0.8.3
pexpect==4.8.0
pickleshare==0.7.5
Pillow==10.0.1
platformdirs==3.11.0
prompt-toolkit==3.0.39
psutil==5.9.5
ptyprocess==0.7.0
pure-eval==0.2.2
Pygments==2.16.1
pyparsing==3.1.1
python-dateutil==2.8.2
pytz==2023.3.post1
pyzmq==25.1.1
scipy==1.11.3
seaborn==0.12.0
six==1.16.0
stack-data==0.6.3
statannotations==0.6.0
tornado==6.3.3
traitlets==5.11.1
wcwidth==0.2.8

Are you sure that you set the versions correctly? What's the output of pip freeze? It's also hard to know why it isn't working for you because you haven't added a minimal working example to your issue.

I was also able to run the following code that makes a violinplot:
`%matplotlib inline
import matplotlib.pyplot as plt
import seaborn as sns
from statannotations.Annotator import Annotator

sns.set(style="whitegrid")

df = sns.load_dataset("tips")

x = "day"
y = "total_bill"
order = ["Sun", "Thur", "Fri", "Sat"]
ax = sns.violinplot(data=df, x=x, y=y, order=order)
annot = Annotator(
ax, [("Thur", "Fri"), ("Thur", "Sat"), ("Fri", "Sun")], data=df, plot="violinplot", x=x, y=y, order=order
)
annot.configure(test="Mann-Whitney", text_format="star", loc="outside", verbose=2)
annot.apply_test()
ax, test_results = annot.annotate()
plt.savefig("example_non-hue_outside.png", dpi=300, bbox_inches="tight")`

[noahsprent]

Looking at your issue it seems like you might have tried with Seaborn 0.11.2 and Seaborn 0.13. Have you tried 0.12?

[sjg2203]

I downgraded Seaborn to version 0.12.2 and it worked! Thanks @noahsprent

However, I am now getting this warning which is the reason why it doesn't work anymore with Seaborn 0.13.0
@trevismd, I believe updating your package is more than necessary in order to keep it functional, can it be done?

C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.
  with pd.option_context('mode.use_inf_as_na', True):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.
  with pd.option_context('mode.use_inf_as_na', True):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\seaborn\_oldcore.py:1498: FutureWarning: is_categorical_dtype is deprecated and will be removed in a future version. Use isinstance(dtype, CategoricalDtype) instead
  if pd.api.types.is_categorical_dtype(vector):
C:\ProgramData\miniforge\envs\statenv\lib\site-packages\statannotations\Annotator.py:775: UserWarning: Changing alpha without updating pvalue_thresholds can result in inconsistent plotting results
  warnings.warn("Changing alpha without updating")```

[sjg2203]

Would be great to add support for seaborn 0.13 as it just loomed out as well as pandas 2.1.1

[dany-fu]

I'm having the same issue as #133 with a different categorical plotter. Looks like manually installing seaborn 0.12.2 is the current workaround, I'm just throwing my support behind creating a more robust fix by not accessing seaborn's internal classes.

[agartland]

Agree. I find this package to be quite useful so would be good to get it up to date with pandas and seaborn before they get too far ahead!
@trevismd can you comment on what kind of effort would be required for the update?
Thanks!