No automatic type casting in complete indexing of a large MultiIndex · Issue #18818 · pandas-dev/pandas (original) (raw)

Code Sample, a copy-pastable example if possible

In [2]: pd.MultiIndex.from_product([[1., 2.], range(5000)]).get_loc((1, 0)) Out[2]: 0

In [3]: pd.MultiIndex.from_product([[1., 2.], range(5001)]).get_loc((1., 0)) Out[3]: 0

In [4]: pd.MultiIndex.from_product([[1., 2.], range(5001)]).get_loc(1) Out[4]: slice(0, 5001, None)

In [5]: pd.MultiIndex.from_product([[1., 2.], range(5001)]).get_loc((1, 0))

KeyError Traceback (most recent call last) in () ----> 1 pd.MultiIndex.from_product([[1., 2.], range(5001)]).get_loc((1, 0))

/home/nobackup/repo/pandas/pandas/core/indexes/multi.py in get_loc(self, key, method) 2141 key = _values_from_object(key) 2142 key = tuple(map(_maybe_str_to_time_stamp, key, self.levels)) -> 2143 return self._engine.get_loc(key) 2144 2145 # -- partial selection or non-unique index

/home/nobackup/repo/pandas/pandas/_libs/index.pyx in pandas._libs.index.MultiIndexHashEngine.get_loc (pandas/_libs/index.c:15856)() 645 return algos.pad_object(values, other, limit=limit) 646 --> 647 cpdef get_loc(self, object val): 648 if is_definitely_invalid_key(val): 649 raise TypeError("'{val}' is an invalid key".format(val=val))

/home/nobackup/repo/pandas/pandas/_libs/index.pyx in pandas._libs.index.MultiIndexHashEngine.get_loc (pandas/_libs/index.c:15703)() 654 655 try: --> 656 return self.mapping.get_item(val) 657 except TypeError: 658 raise KeyError(val)

/home/nobackup/repo/pandas/pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.MultiIndexHashTable.get_item (pandas/_libs/hashtable.c:24621)() 1446 return False 1447 -> 1448 cpdef get_item(self, object key): 1449 cdef: 1450 khiter_t k

/home/nobackup/repo/pandas/pandas/_libs/hashtable_class_helper.pxi in pandas._libs.hashtable.MultiIndexHashTable.get_item (pandas/_libs/hashtable.c:24576)() 1460 return loc 1461 else: -> 1462 raise KeyError(key) 1463 1464 cpdef set_item(self, object key, Py_ssize_t val):

KeyError: (1, 0)

Problem description

Related to #18519 - this is an inherent limit of the current design of the engine for large MultiIndexes, and an improved MultiIndexEngine should solve this too.

Expected Output

Output of pd.show_versions()

INSTALLED VERSIONS

commit: b5f1e71
python: 3.5.3.final.0
python-bits: 64
OS: Linux
OS-release: 4.9.0-4-amd64
machine: x86_64
processor:
byteorder: little
LC_ALL: None
LANG: it_IT.UTF-8
LOCALE: it_IT.UTF-8

pandas: 0.22.0.dev0+371.gb5f1e716d
pytest: 3.2.3
pip: 9.0.1
setuptools: 36.7.0
Cython: 0.25.2
numpy: 1.12.1
scipy: 0.19.0
pyarrow: None
xarray: None
IPython: 6.2.1
sphinx: 1.5.6
patsy: 0.4.1
dateutil: 2.6.1
pytz: 2017.2
blosc: None
bottleneck: 1.2.0dev
tables: 3.3.0
numexpr: 2.6.1
feather: 0.3.1
matplotlib: 2.0.0
openpyxl: 2.3.0
xlrd: 1.0.0
xlwt: 1.3.0
xlsxwriter: 0.9.6
lxml: 4.1.1
bs4: 4.5.3
html5lib: 0.999999999
sqlalchemy: 1.0.15
pymysql: None
psycopg2: None
jinja2: 2.10
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: 0.2.1