Anaconda 5.0.1
Oct 25, 2017
  • R has been updated to version 3.4.2. All R packages (including RStudio) have been rebuilt to be compatible with the new Anaconda 5.0 compilers.
  • Updated many packages, including Python, Numpy, Spyder, Navigator, and Bokeh.
  • The MKL library load path has been modified to address issue for Julia users.
  • Fixed an OpenSSL issue with WSL on Windows.
  • Fixed Anaconda Installer Configuration (AIC) feature for Unix installers.
  • Re-enabled spaces in installation paths on Windows (temporarily disabled in 5.0.0).

Major package updates

  • anaconda-navigator 1.6.8 -> 1.6.9
  • bokeh 0.12.7 -> 0.12.10
  • conda 4.3.27 -> 4.3.30
  • conda-build 3.0.22 -> 3.0.27
  • dask 0.15.2 -> 0.15.3
  • matplotlib 2.0.2 -> 2.1.0
  • numpy 1.13.1 -> 1.13.3
  • python 2.7.13 -> 2.7.14
  • python 3.6.2 -> 3.6.3
  • scikit-learn 0.19.0 -> 0.19.1
  • spyder 3.2.3 -> 3.2.4
Anaconda 5.0.0.1
Oct 02, 2017
  • Fixes Python & C compiler fallback path for all cases on x86/x86_64 Linux. Without this fix, people were required to use our new compilers, which is not something we want to enforce at this time. This was affecting travis-ci builds and pip installs of packages that require compilation for extensions.
Anaconda 5.0.0
Sep 26, 2017
  • Over 100 packages updated and added. MKL is updated to 2018.0.0. JupyterLab alpha preview 0.27.0 is included.
  • All new compilers on macOS and Linux, giving substantial security and performance improvements.
  • Where possible, all build recipes use conda-forge as a base, using AnacondaRecipes .
  • A new channel, pkgs/main, has been added to defaults. The new channel is given top priority within defaults and holds packages built with the new compiler stack.
  • Continuum Analytics has been renamed to Anaconda, Inc. See this blog post for more.
  • Spaces are no longer allowed in the installation path on Windows.
  • Transitioned to more flexible dependency pinning of numpy packages, giving wider ranges of compatibility.

Major package updates

  • anaconda-client 1.6.3 -> 1.6.5
  • anaconda-navigator 1.6.2 -> 1.6.8
  • anaconda-project 0.6.0 -> 0.8.0
  • bokeh 0.12.5 -> 0.12.7
  • conda 4.3.21 -> 4.3.27
  • dask 0.14.3 -> 0.15.2
  • jupyter_client 5.0.1 -> 5.1.0
  • jupyter_console 5.1.0 -> 5.2.0
  • numba 0.33.0 -> 0.35.0
  • numpy 1.12.1 -> 1.13.1
  • pandas 0.20.1 -> 0.20.3
  • pillow 4.1.1 -> 4.2.1
  • requests 2.14.2 -> 2.18.4
  • scikit-learn 0.18.1 -> 0.19.0
  • scipy 0.19.0 -> 0.19.1
  • seaborn 0.7.1 -> 0.8.0
  • spyder 3.1.4 -> 3.2.3
  • sympy 1.0 -> 1.1.1
  • conda-build 3.0.22
  • jupyterlab 0.27.0
  • jupyterlab_launcher 0.4.0
Anaconda 4.4.0
May 31, 2017

Anaconda 4.4 was released on May 31, 2017 and includes the following:

  • Support added for the “ppc64le” machine type, for the POWER8 LE architecture used by IBM Power Systems and OpenPOWER servers.
  • On Windows, the PATH environment variable is no longer changed by default, as this can cause trouble with other software. Instead, use Anaconda Navigator or the Anaconda Prompt in the Start Menu under “Anaconda” to use Anaconda software. If a user does choose to change the PATH variable, Anaconda is no longer appended to the PATH in system mode, and is now always added to the front of PATH in either system mode or user mode.
  • Python 3.5 is updated from 3.5.2 to 3.5.3 and Python 3.6 from 3.6.0 to 3.6.1. Anaconda 4.4 supports Python 2.7, 3.5, and 3.6. Anaconda 4.3 was the last release to support Python 3.4.
  • Minimum supported version of CentOS is now CentOS 6. Anaconda 4.3 was the last release to support CentOS 5.
  • Applied pycrypto patch for CVE-2013-7439.
  • Improved cp_acp support for install paths with non-ASCII characters on Windows.
  • conda is updated from 4.3.14 to 4.3.21.
  • Navigator is updated from 1.5.0 to 1.6.2.
  • Project is updated from 0.4.1 to 0.6.0.
  • Added distributed and pyodbc to the installers.
  • Updated EULA.
  • Conda packages with “mkl” in the package name now contain a file license.txt with a copy of the Intel Simplified Software License that applies to the Intel Math Kernel Library (MKL).
  • Over 90 packages are updated or added.

Highlights:

  • add support for the ppc64le (POWER8 LE used by IBM Power Systems and OpenPOWER servers) machine types

Other changes:

  • On Windows, the PATH environment variable is no longer changed by default, as this can cause trouble with other software. The recommended approach is to instead use Anaconda Navigator or the Anaconda Prompt (located in the Start Menu under “Anaconda”) when you wish to use Anaconda software. Also, Anaconda will always be added to the front of PATH, for either system or user mode. (Previously it was appended to the system path.)
  • improve cp_acp support for install path on Windows
  • updated 80 packages in the installer (and their dependencies)
  • added distributed and pyodbc to the installers
  • apply pycrypto patch for CVE-2013-7439
  • end support for CentOS 5. CentOS 6 is now the minimum supported version.

Updates:

  • anaconda-client from 1.6.0 to 1.6.3
  • bokeh from 0.12.4 to 0.12.5
  • dask from 0.13.0 to 0.14.3
  • ipython from 5.1.0 to 5.3.0
  • jupyter_client from 4.4.0 to 5.0.1
  • jupyter_console from 5.0.0 to 5.1.0
  • jupyter_core from 4.2.1 to 4.3.0
  • matplotlib from 2.0.0 to 2.0.2
  • numba from 0.30.1 to 0.33.0
  • numpy from 1.11.3 to 1.12.1
  • pandas from 0.19.2 to 0.20.1
  • pillow from 4.0.0 to 4.1.1
  • python 3.5 from 3.5.2 to 3.5.3
  • python 3.6 from 3.6.0 to 3.6.1
  • requests from 2.12.4 to 2.14.2
  • scikit-image from 0.12.3 to 0.13.0
  • scipy from 0.18.1 to 0.19.0
  • spyder from 3.1.2 to 3.1.4
  • statsmodels from 0.6.1 to 0.8.0
Anaconda 4.3.1
Mar 10, 2017

This patch release fixes problems with Anaconda Navigator not starting correctly on some versions of Mac OS X when using the GUI installers.

Fixes:

  • removed creation of ~/.continuum folder during install process on all platforms
  • fixed ‘//’ showing up in prefix when installing system wide on Mac OS using the GUI installer
  • fixed OpenSSL not being installable into a path which contains spaces
  • allow Unicode characters in install path on Windows (cp_acp fix)

Updates:

  • anaconda-navigator from 1.4.3 to 1.5.0
  • conda from 4.3.8 to 4.3.14

Added:

  • anaconda-project 0.4.1
Anaconda 4.3.0.1
Feb 03, 2017

In this “mirco” patch release, we fixed a problem with the Windows installers which was causing problems with Qt applications when the install prefix exceeds 30 characters. No new Anaconda meta-packages correspond to this release (only new Windows installers).

Anaconda 4.3.0
jan 31, 2017

Highlights:

  • The Anaconda3 installers are based on Python 3.6. Anaconda 4.3 supports Python 2.7, 3.4, 3.5 and 3.6. Anaconda 4.3 will be the last release which supports Python 3.4. We will discontinue regular Python 3.4 package updates in the next release.
  • The Intel Math Kernel Library (MKL) is updated from 11.3.3 to 2017.0.1.
  • Over 90 packages are updated.
  • seaborn is now installed by default.

Other changes:

  • Updates jpeg and libpng to increase compatibility with conda-forge.
  • Warns about possible errors if installing on Windows into an install path with spaces, and does not allow installation if the install path contains unicode characters.
  • Fixes many Windows menu uninstallation issues and some other often reported uninstallation issues on Windows.
  • Anaconda 4.2 is the last release that supports macOS 10.7 and macOS 10.8. Anaconda 4.3 supports macOS versions from 10.9 through the current version 10.12.
  • conda-build, anaconda-clean and the Jupyter Notebook extensions are no longer installed by default but can be installed with a single conda command.

Major package updates:

  • anaconda-client from 1.5.1 to 1.6.0
  • anaconda-navigator from 1.3.1 to 1.4.3
  • bokeh from 0.12.2 to 0.12.4
  • conda from 4.2.9 to 4.3.8
  • dask from 0.11.0 to 0.13.0
  • jupyter_core from 4.2.0 to 4.2.1
  • matplotlib from 1.5.3 to 2.0.0
  • mkl from 11.3.3 to 2017.0.1
  • numba from 0.28.1 to 0.30.1
  • numpy from 1.11.1 to 1.11.3
  • pandas from 0.18.1 to 0.19.2
  • pillow from 3.3.1 to 4.0.0
  • python from 2.7.12 to 2.7.13
  • requests from 2.11.1 to 2.12.4
  • scikit-learn from 0.17.1 to 0.18.1
  • spyder from 3.0.0 to 3.1.2
  • seaborn 0.7.1

Was this page helpful?