OptionError: 'Pattern matched multiple keys' for scv.get_df()

[hyjforesight]

Hello scVelo,
I met a OptionError: 'Pattern matched multiple keys' when running scv.get_df() of the tutorial 'RNA Velocity Basics'.
Please see the code below.
Could you please help me with this issue?
Thanks!
Best,
YJ

adata.uns['neighbors']['distances'] = adata.obsp['distances']
adata.uns['neighbors']['connectivities'] = adata.obsp['connectivities']
scv.tl.paga(adata, groups='clusters', vkey='velocity', use_time_prior="velocity_pseudotime")

df = scv.get_df(adata, keys='paga/transitions_confidence', precision=2).T    # no error until this step
OptionError                               Traceback (most recent call last)
Input In [58], in <cell line: 1>()
----> 1 df = scv.get_df(adata, keys='paga/transitions_confidence', precision=2).T
      2 df.style.background_gradient(cmap='Blues').format('{:.2g}')

File ~\.conda\envs\new38\lib\site-packages\scvelo\core\_anndata.py:196, in get_df(data, keys, layer, index, columns, sort_values, dropna, precision)
    164 """Get dataframe for a specified adata key.
    165 
    166 Return values for specified key
   (...)
    192     A dataframe.
    193 """
    195 if precision is not None:
--> 196     pd.set_option("precision", precision)
    198 if isinstance(data, AnnData):
    199     keys, keys_split = (
    200         keys.split("*") if isinstance(keys, str) and "*" in keys else (keys, None)
    201     )

File ~\.conda\envs\new38\lib\site-packages\pandas\_config\config.py:256, in CallableDynamicDoc.__call__(self, *args, **kwds)
    255 def __call__(self, *args, **kwds):
--> 256     return self.__func__(*args, **kwds)

File ~\.conda\envs\new38\lib\site-packages\pandas\_config\config.py:149, in _set_option(*args, **kwargs)
    146     raise TypeError(f'_set_option() got an unexpected keyword argument "{kwarg}"')
    148 for k, v in zip(args[::2], args[1::2]):
--> 149     key = _get_single_key(k, silent)
    151     o = _get_registered_option(key)
    152     if o and o.validator:

File ~\.conda\envs\new38\lib\site-packages\pandas\_config\config.py:116, in _get_single_key(pat, silent)
    114     raise OptionError(f"No such keys(s): {repr(pat)}")
    115 if len(keys) > 1:
--> 116     raise OptionError("Pattern matched multiple keys")
    117 key = keys[0]
    119 if not silent:

OptionError: 'Pattern matched multiple keys'

Error output

# paste the error output here, if applicable

scvelo==0.2.4  scanpy==1.9.1  anndata==0.7.5  loompy==3.0.6  numpy==1.21.5  scipy==1.8.0  matplotlib==3.5.1  sklearn==1.0.2  pandas==1.4.1 

[WeilerP]

@hyjforesight, seems like an issue is related to Pandas not scVelo. I'm pretty sure the issue is that you are using older arguments for Pandas options (in your case precision). Starting from Pandas==1.4 some of the arguments are replaced by longer forms (display.precision in your case). See here. Please let me know if this resolves the problem.

[hyjforesight]

hello @WeilerP
Thanks for the response.

Because the parameters for scVelo is

scv.get_df(
    data: anndata._core.anndata.AnnData,
    keys: Union[str, List[str], NoneType] = None,
    layer: Union[str, NoneType] = None,
    index: List = None,
    columns: List = None,
    sort_values: bool = None,
    dropna: Literal['all', 'any'] = 'all',
    precision: int = None,
)

I set the argument precision=2. Should I replace precision as display.precision? But I got the errors as below:

df = scv.get_df(data=adata, keys='paga/transitions_confidence', display.precision=2).T
SyntaxError: expression cannot contain assignment, perhaps you meant "=="?

df = scv.get_df(data=adata, keys='paga/transitions_confidence', display.precision==2).T
SyntaxError: positional argument follows keyword argument

[WeilerP]

Ah sorry, I thought it was a keyword directly passed to Pandas. Should have actually checked the code first. I opened a PR but unit tests are failing ATM. Will have to investigate once I find the time.

[WeilerP]

Fixed in #907.