Skip to content
Simple graphical tabular data viewer
Branch: master
Clone or download
Type Name Latest commit message Commit time
Failed to load latest commit information.
bin Favor python3 by default Oct 23, 2017
sample Mention/test Blaze support. Jul 30, 2015
tests models: special-case list of bytes/strings May 25, 2016
.gitignore Update classifiers and ignores Jun 17, 2016
.travis.yml Revert "travis: alternate method to remove default build" Apr 6, 2017
LICENSE.txt Bump copyright dates Apr 15, 2016
NEWS.rst Bump version and release notes Dec 12, 2016
README.rst Drop setuptools-git Oct 23, 2017


gtabview: a simple graphical tabular data viewer

Graphical counterpart to tabview, a simple tabular data viewer that can be used both stand-alone and as a Python module for various files and Python/Pandas/NumPy data structures.

Stand-alone usage

gtabview reads most text tabular data formats automatically:

gtabview data.csv
gtabview data.txt

If xlrd is installed, Excel files can be read directly:

gtabview file.xls[x]

When Blaze is also installed, any Blaze source can be used by specifying a supported URI on the command line:

gtabview file://dataset.hdf5
gtabview file://dataset.json
gtabview sqlite://file.db::table
gtabview postgresql://host.domain/db_name::table

The database URL syntax is inherited from SQLAlchemy, so refer to SQLAlchemy's database URLs for a detailed reference.

Usage as a module

gtabview.view() can be used to display simple Python types directly in tabulated form:

from gtabview import view

# view a file

# view a list
view([1, 2, 3])

# view a dict (by columns)
view({'a': [1, 2, 3], 'b': [4, 5, 6], 'c': [7, 8, 9]})

# view a dict (by rows)
view({'a': [1, 2, 3], 'b': [4, 5, 6], 'c': [7, 8, 9]}, transpose=True)

# view a simple list of lists
view([[1, 2, 3], [4, 5, 6], [7, 8, 9]])

# view a simple list of lists (with headers)
view([['a', 'b', 'c'], [1, 2, 3], [4, 5, 6], [7, 8, 9]], hdr_rows=1)

gtabview includes native support for NumPy and all features of Pandas' DataFrames, such as MultiIndexes and level names:

from gtabview import view

# numpy arrays up to two dimensions are supported
import numpy as np
view(np.array([[1, 2, 3], [4, 5, 6]]))

# view a DataFrame/Series/Panel
import pandas as pd
df = pd.DataFrame([[1, 2, 3], [4, 5, 6]],
                  columns=['a', 'b', 'c'], index=['x', 'y'])

Blaze can also be used directly as a data source, either explicitly or implicitly through an URI:

from gtabview import view
import blaze as bz

iris = bz.Data('sqlite:///blaze/examples/data/iris.db::iris')


gtabview is designed to integrate correctly with matplotlib. If you're using gtabview with matplotlib either directly or indirectly (for example, using the Pandas visualization API or Seaborn), be sure to include matplotlib first to correctly initialize gtabview.

gtabview will also use matplotlib's interactive setting to determine the default behavior of the data window: when interactive, calls to view() will not block, and will keep recycling the same window.

To use gtabview in a Python Notebook with inline graphics, you'll probably want to force the detached behavior. In the first cell of your notebook, initialize both gtabview and matplotlib as follows:

import gtabview
gtabview.DETACH = True
from gtabview import view
%matplotlib inline

When using view, a separate data window will show. The window can be kept around or closed, but will only be refreshed when evaluating the cell again.

Requirements and installation

gtabview is available directly on the Python Package Index.

gtabview requires:

  • Python 3 or Python 2
  • PyQt5, PyQt4 or PySide
  • setuptools (install-only)

Under Debian/Ubuntu, install the required dependencies with:

sudo apt-get install python3 python3-pyqt5
sudo apt-get install python3-setuptools

Then download and install simply via pip:

pip install gtabview

Install xlrd if reading Excel files directly is desired, and optionally Blaze for interacting with other/scientific data formats:

pip install xlrd
pip install blaze


gtabview is distributed under the MIT license (see LICENSE.txt)
Copyright(c) 2014-2017: wave++ "Yuri D'Elia" <>
Copyright(c) 2014-2015: Scott Hansen <>
You can’t perform that action at this time.