BUG: np.array @ pd.Series swaps arguments

[adbull]

Code Sample, a copy-pastable example if possible

Input:

np.zeros((1, 2)) @ pd.Series(np.zeros(2))

Output:

  File "<ipython-input-83-3148232e4345>", line 1, in <module>
    np.zeros((1, 2)) @ pd.Series(np.zeros(2))
  File "pandas/core/series.py", line 2061, in __rmatmul__
    return self.dot(other)
  File "pandas/core/series.py", line 2043, in dot
    (lvals.shape, rvals.shape))
Exception: Dot product shape mismatch, (2,) vs (1, 2)

Problem description

Pandas v0.23.0 (#19035) adds support for expressions np.array @ pd.Series, but the support is implemented incorrectly, and instead evaluates pd.Series @ np.array, which is different. This is inconsistent with the behaviour of @ when applied to two arrays, two pandas objects, or an array and a DataFrame.

Expected Output

array([ 0.])

Output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 3.6.5.final.0
python-bits: 64
OS: Linux
OS-release: 4.14.14-200.fc26.x86_64
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: C
LANG: C
LOCALE: None.None

pandas: 0.23.1
pytest: 3.6.0
pip: 10.0.1
setuptools: 39.2.0
Cython: 0.28.3
numpy: 1.13.3
scipy: 0.19.1
pyarrow: None
xarray: None
IPython: 4.2.1
sphinx: None
patsy: 0.5.0
dateutil: 2.7.3
pytz: 2018.4
blosc: None
bottleneck: 1.2.1
tables: 3.4.3
numexpr: 2.6.2
feather: None
matplotlib: 2.2.2
openpyxl: None
xlrd: 1.1.0
xlwt: None
xlsxwriter: None
lxml: None
bs4: 4.6.0
html5lib: 1.0.1
sqlalchemy: 1.2.8
pymysql: None
psycopg2: None
jinja2: 2.10
s3fs: None
fastparquet: 0.1.5
pandas_gbq: None
pandas_datareader: None

[gfyoung]

cc @TomAugspurger

[TomAugspurger]

@bnaul

[minggli]

happy to give it a try.

[gfyoung]

@minggli : Go for it!

[minggli]

raised a PR to address this issue.

Series.__matmul__ is called when Series * something
Series.__rmatmul__ is called when something @ Series and that something doesn't have __matmul__ method