Frequently Asked Questions (FAQ) — pandas 0.18.1 documentation (original) (raw)

DataFrame memory usage¶

As of pandas version 0.15.0, the memory usage of a dataframe (including the index) is shown when accessing the info method of a dataframe. A configuration option, display.memory_usage (see Options and Settings), specifies if the dataframe’s memory usage will be displayed when invoking the df.info() method.

For example, the memory usage of the dataframe below is shown when calling df.info():

In [1]: dtypes = ['int64', 'float64', 'datetime64[ns]', 'timedelta64[ns]', ...: 'complex128', 'object', 'bool'] ...:

In [2]: n = 5000

In [3]: data = dict([ (t, np.random.randint(100, size=n).astype(t)) ...: for t in dtypes]) ...:

In [4]: df = pd.DataFrame(data)

In [5]: df['categorical'] = df['object'].astype('category')

In [6]: df.info() <class 'pandas.core.frame.DataFrame'> RangeIndex: 5000 entries, 0 to 4999 Data columns (total 8 columns): bool 5000 non-null bool complex128 5000 non-null complex128 datetime64[ns] 5000 non-null datetime64[ns] float64 5000 non-null float64 int64 5000 non-null int64 object 5000 non-null object timedelta64[ns] 5000 non-null timedelta64[ns] categorical 5000 non-null category dtypes: bool(1), category(1), complex128(1), datetime64ns, float64(1), int64(1), object(1), timedelta64ns memory usage: 284.1+ KB

The + symbol indicates that the true memory usage could be higher, because pandas does not count the memory used by values in columns withdtype=object.

New in version 0.17.1.

Passing memory_usage='deep' will enable a more accurate memory usage report, that accounts for the full usage of the contained objects. This is optional as it can be expensive to do this deeper introspection.

In [7]: df.info(memory_usage='deep') <class 'pandas.core.frame.DataFrame'> RangeIndex: 5000 entries, 0 to 4999 Data columns (total 8 columns): bool 5000 non-null bool complex128 5000 non-null complex128 datetime64[ns] 5000 non-null datetime64[ns] float64 5000 non-null float64 int64 5000 non-null int64 object 5000 non-null object timedelta64[ns] 5000 non-null timedelta64[ns] categorical 5000 non-null category dtypes: bool(1), category(1), complex128(1), datetime64ns, float64(1), int64(1), object(1), timedelta64ns memory usage: 401.2 KB

By default the display option is set to True but can be explicitly overridden by passing the memory_usage argument when invoking df.info().

The memory usage of each column can be found by calling the memory_usagemethod. This returns a Series with an index represented by column names and memory usage of each column shown in bytes. For the dataframe above, the memory usage of each column and the total memory usage of the dataframe can be found with the memory_usage method:

In [8]: df.memory_usage() Out[8]: Index 72 bool 5000 complex128 80000 datetime64[ns] 40000 float64 40000 int64 40000 object 40000 timedelta64[ns] 40000 categorical 5800 dtype: int64

total memory usage of dataframe

In [9]: df.memory_usage().sum() Out[9]: 290872

By default the memory usage of the dataframe’s index is shown in the returned Series, the memory usage of the index can be suppressed by passing the index=False argument:

In [10]: df.memory_usage(index=False) Out[10]: bool 5000 complex128 80000 datetime64[ns] 40000 float64 40000 int64 40000 object 40000 timedelta64[ns] 40000 categorical 5800 dtype: int64

The memory usage displayed by the info method utilizes thememory_usage method to determine the memory usage of a dataframe while also formatting the output in human-readable units (base-2 representation; i.e., 1KB = 1024 bytes).

See also Categorical Memory Usage.

Byte-Ordering Issues¶

Occasionally you may have to deal with data that were created on a machine with a different byte order than the one on which you are running Python. To deal with this issue you should convert the underlying NumPy array to the native system byte order before passing it to Series/DataFrame/Panel constructors using something similar to the following:

In [11]: x = np.array(list(range(10)), '>i4') # big endian

In [12]: newx = x.byteswap().newbyteorder() # force native byteorder

In [13]: s = pd.Series(newx)

See the NumPy documentation on byte order for more details.

Visualizing Data in Qt applications¶

Warning

The qt support is deprecated and will be removed in a future version. We refer users to the external package pandas-qt.

There is experimental support for visualizing DataFrames in PyQt4 and PySide applications. At the moment you can display and edit the values of the cells in the DataFrame. Qt will take care of displaying just the portion of the DataFrame that is currently visible and the edits will be immediately saved to the underlying DataFrame

To demonstrate this we will create a simple PySide application that will switch between two editable DataFrames. For this will use the DataFrameModel class that handles the access to the DataFrame, and the DataFrameWidget, which is just a thin layer around the QTableView.

import numpy as np import pandas as pd from pandas.sandbox.qtpandas import DataFrameModel, DataFrameWidget from PySide import QtGui, QtCore

Or if you use PyQt4:

from PyQt4 import QtGui, QtCore

class MainWidget(QtGui.QWidget): def init(self, parent=None): super(MainWidget, self).init(parent)

    # Create two DataFrames
    self.df1 = pd.DataFrame(np.arange(9).reshape(3, 3),
                            columns=['foo', 'bar', 'baz'])
    self.df2 = pd.DataFrame({
            'int': [1, 2, 3],
            'float': [1.5, 2.5, 3.5],
            'string': ['a', 'b', 'c'],
            'nan': [np.nan, np.nan, np.nan]
        }, index=['AAA', 'BBB', 'CCC'],
        columns=['int', 'float', 'string', 'nan'])

    # Create the widget and set the first DataFrame
    self.widget = DataFrameWidget(self.df1)

    # Create the buttons for changing DataFrames
    self.button_first = QtGui.QPushButton('First')
    self.button_first.clicked.connect(self.on_first_click)
    self.button_second = QtGui.QPushButton('Second')
    self.button_second.clicked.connect(self.on_second_click)

    # Set the layout
    vbox = QtGui.QVBoxLayout()
    vbox.addWidget(self.widget)
    hbox = QtGui.QHBoxLayout()
    hbox.addWidget(self.button_first)
    hbox.addWidget(self.button_second)
    vbox.addLayout(hbox)
    self.setLayout(vbox)

def on_first_click(self):
    '''Sets the first DataFrame'''
    self.widget.setDataFrame(self.df1)

def on_second_click(self):
    '''Sets the second DataFrame'''
    self.widget.setDataFrame(self.df2)

if name == 'main': import sys

# Initialize the application
app = QtGui.QApplication(sys.argv)
mw = MainWidget()
mw.show()
app.exec_()