unibench show_results - ValueError: too many inputs

[LudovicArnould1]

Thank you for your amazing work.

After I installed your library using pip install git+ ..., I received a ValueError: too many inputs when running unibench show_results.

Apparently there is a limitation of the number of conditions passed in pandas.query() due to the numexpr package where a NPY_MAXARGS argument is set (related issue).

Maybe it can be solved by upgrading numpy to 2.1 but I did not try. So I slightly changed the query method from the OutputHandler Class by using a mask and applying the isin() method to efficiently look for the list/df value matching.

    def query(self, df=None, **kwargs):
        if df is None:
            df = self._model_csv
        if len(kwargs) == 0:
            return df

        mask = pd.Series([True] * len(df))

        for k, v in kwargs.items():
            if isinstance(v, list):
                mask &= df[k].isin(v)
            else:
                mask &= (df[k] == v)

        return df[mask]

Then unibench show_results worked properly.

The full Traceback :

Traceback (most recent call last):
  File "/home//anaconda3/bin/unibench", line 8, in <module>
    sys.exit(run())
             ^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/unibench/evaluator.py", line 321, in run
    fire.Fire(Evaluator)
  File "/home//anaconda3/lib/python3.11/site-packages/fire/core.py", line 143, in Fire
    component_trace = _Fire(component, args, parsed_flag_args, context, name)
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/fire/core.py", line 477, in _Fire
    component, remaining_args = _CallAndUpdateTrace(
                                ^^^^^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/fire/core.py", line 693, in _CallAndUpdateTrace
    component = fn(*varargs, **kwargs)
                ^^^^^^^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/unibench/evaluator.py", line 172, in show_results
    self.outputhandler.print_dataframe(
  File "/home//anaconda3/lib/python3.11/site-packages/unibench/output.py", line 188, in print_dataframe
    df = self.query(df=self._aggregate, **kwargs)
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/unibench/output.py", line 143, in query
    return df.query(expr)
           ^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/pandas/core/frame.py", line 4599, in query
    res = self.eval(expr, **kwargs)
          ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/pandas/core/frame.py", line 4725, in eval
    return _eval(expr, inplace=inplace, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/pandas/core/computation/eval.py", line 357, in eval
    ret = eng_inst.evaluate()
          ^^^^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/pandas/core/computation/engines.py", line 81, in evaluate
    res = self._evaluate()
          ^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/pandas/core/computation/engines.py", line 121, in _evaluate
    return ne.evaluate(s, local_dict=scope)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/numexpr/necompiler.py", line 973, in evaluate
    return re_evaluate(local_dict=local_dict, _frame_depth=_frame_depth)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home//anaconda3/lib/python3.11/site-packages/numexpr/necompiler.py", line 1004, in re_evaluate
    return compiled_ex(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
ValueError: too many inputs

Config :
Ubuntu 22.04.4
Pandas: 2.1.4
Python: 3.11.7
Pip: 23.3.1
Numpy : 1.26.4

[haideraltahan]

Hi @LudovicArnould1

Thank you for pointing that out, I will push a new release with a fix!