.. currentmodule:: pandas
.. ipython:: python :suppress: import numpy as np np.random.seed(123456) np.set_printoptions(precision=4, suppress=True) import pandas as pd pd.options.display.max_rows = 15
This section will focus on downstream applications of pandas.
The Apache Parquet format provides key-value metadata at the file and column level, stored in the footer of the Parquet file:
5: optional list<KeyValue> key_value_metadata
where KeyValue
is
struct KeyValue {
1: required string key
2: optional string value
}
So that a pandas.DataFrame
can be faithfully reconstructed, we store a
pandas
metadata key in the FileMetaData
with the value stored as :
{'index_columns': ['__index_level_0__', '__index_level_1__', ...],
'column_indexes': [<ci0>, <ci1>, ..., <ciN>],
'columns': [<c0>, <c1>, ...],
'pandas_version': $VERSION}
Here, <c0>
/<ci0>
and so forth are dictionaries containing the metadata
for each column. This has JSON form:
{'name': column_name,
'pandas_type': pandas_type,
'numpy_type': numpy_type,
'metadata': metadata}
pandas_type
is the logical type of the column, and is one of:
- Boolean:
'bool'
- Integers:
'int8', 'int16', 'int32', 'int64', 'uint8', 'uint16', 'uint32', 'uint64'
- Floats:
'float16', 'float32', 'float64'
- Date and Time Types:
'datetime', 'datetimetz'
,'timedelta'
- String:
'unicode', 'bytes'
- Categorical:
'categorical'
- Other Python objects:
'object'
The numpy_type
is the physical storage type of the column, which is the
result of str(dtype)
for the underlying NumPy array that holds the data. So
for datetimetz
this is datetime64[ns]
and for categorical, it may be
any of the supported integer categorical types.
The metadata
field is None
except for:
datetimetz
:{'timezone': zone, 'unit': 'ns'}
, e.g.{'timezone', 'America/New_York', 'unit': 'ns'}
. The'unit'
is optional, and if omitted it is assumed to be nanoseconds.categorical
:{'num_categories': K, 'ordered': is_ordered, 'type': $TYPE}
- Here
'type'
is optional, and can be a nested pandas type specification here (but not categorical)
- Here
unicode
:{'encoding': encoding}
- The encoding is optional, and if not present is UTF-8
object
:{'encoding': encoding}
. Objects can be serialized and stored inBYTE_ARRAY
Parquet columns. The encoding can be one of:'pickle'
'msgpack'
'bson'
'json'
timedelta
:{'unit': 'ns'}
. The'unit'
is optional, and if omitted it is assumed to be nanoseconds. This metadata is optional altogether
For types other than these, the 'metadata'
key can be
omitted. Implementations can assume None
if the key is not present.
As an example of fully-formed metadata:
{'index_columns': ['__index_level_0__'],
'column_indexes': [
{'name': None,
'pandas_type': 'string',
'numpy_type': 'object',
'metadata': None}
],
'columns': [
{'name': 'c0',
'pandas_type': 'int8',
'numpy_type': 'int8',
'metadata': None},
{'name': 'c1',
'pandas_type': 'bytes',
'numpy_type': 'object',
'metadata': None},
{'name': 'c2',
'pandas_type': 'categorical',
'numpy_type': 'int16',
'metadata': {'num_categories': 1000, 'ordered': False}},
{'name': 'c3',
'pandas_type': 'datetimetz',
'numpy_type': 'datetime64[ns]',
'metadata': {'timezone': 'America/Los_Angeles'}},
{'name': 'c4',
'pandas_type': 'object',
'numpy_type': 'object',
'metadata': {'encoding': 'pickle'}},
{'name': '__index_level_0__',
'pandas_type': 'int64',
'numpy_type': 'int64',
'metadata': None}
],
'pandas_version': '0.20.0'}
.. versionadded:: 0.23.0
Warning
Support for custom array types is experimental.
Sometimes the NumPy type system isn't rich enough for your needs. Pandas has
made a few extensions internally (e.g. Categorical
). While this has worked
well for pandas, not all custom data types belong in pandas itself.
Pandas defines an interface for custom arrays. Arrays implementing this
interface will be stored correctly in Series
or DataFrame
. The ABCs
that must be implemented are
- :class:`ExtensionDtype` A class describing your data type itself. This is
similar to a
numpy.dtype
. - :class:`ExtensionArray`: A container for your data.
Throughout this document, we'll use the example of storing IPv6 addresses. An
IPv6 address is 128 bits, so NumPy doesn't have a native data type for it. We'll
model it as a structured array with two uint64
fields, which together
represent the 128-bit integer that is the IP Address.
This should describe your data type. The most important fields are name
and
base
:
class IPv6Type(ExtensionDtype):
name = 'IPv6'
base = np.dtype([('hi', '>u8'), ('lo', '>u8')])
type = IPTypeType
kind = 'O'
fill_value = np.array([(0, 0)], dtype=base)
base
describe the underlying storage of individual items in your array.
TODO: is this true? Or does .base
refer to the original memory this
is a view on? Different meanings for np.dtype.base
vs. np.ndarray.base
?
In our IPAddress case, we're using a NumPy structured array with two fields.
This is the actual array container for your data, similar to a Categorical
,
and requires the most work to implement correctly. pandas makes no assumptions
about how you store the data. You're free to use NumPy arrays or PyArrow
arrays, or even just Python lists. That said, several of the methods required by
the interface expect NumPy arrays as the return value.
dtype
: Should be an instance of your customExtensionType
formtting_values(self)
: Used for printing Series and DataFrameconcat_same_type(concat)
: Used in :func:`pd.concat`