ANN: QuTiP 4.1.0

Paul Nation

unread,

Mar 11, 2017, 6:05:20 PM3/11/17

to QuTiP: Quantum Toolbox in Python

We are pleased to announce that QuTiP 4.1 was released on March 10th 2017. The major highlights are given below:

- MAJOR FEATURE: QuTiP now works for Python 3.5+ on Windows using Visual Studio 2015.

- MAJOR FEATURE: Can now use interpolating cubic splines as time-dependent coefficients.

- MAJOR FEATURE: Sparse matrix - vector multiplication now can run in parallel using OPENMP.

- Automatic tuning of OPENMP threading threashold.

- Cython and other low level code switched to C++ for MS Windows compatibility.

- Partial trace function is now up to 100x+ faster.

- Hermitian verification now up to 100x+ faster.

- Internal Qobj objects now created up to 60x faster.

- Inplace conversion from COO -> CSR sparse formats (e.g. Memory efficiency improvement.)

- Faster reverse Cuthill-Mckee (10x+) and sparse one and inf norms (10x+).

This release also has a few things that break backward compatibility:

- The Fortran monte carlo solver is no longer bundled with QuTiP. It will be moved to its own installation.

- Only Windows installations using Python 3.5+ and Visual Studio 2015 are officially supported.

- The format of the QuTIP configuration file has changed.

lee Albert

unread,

Mar 13, 2017, 12:45:38 PM3/13/17

to QuTiP: Quantum Toolbox in Python

Did you cancel the --with-f90mc option?
When i compile qutip --with-f90mc error info appeared as `error: option --with-f90mc not recognized ` mcsolver_f90 disappeared.
could you give me more information of 90mc? thanks

在 2017年3月12日星期日 UTC+8上午7:05:20，Paul Nation写道：

Paul Nation

unread,

Mar 13, 2017, 1:36:46 PM3/13/17

to QuTiP Group

It has been removed as it is no longer maintained. It is being moved to its own installer.

-P

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

lee Albert

unread,

Mar 13, 2017, 10:16:54 PM3/13/17

to QuTiP: Quantum Toolbox in Python

Thank you, If i really want to install mcf90, would you give me a method? my platform is Fedora 25


QuTiP: Quantum Toolbox in Python

Copyright (c) 2011 and later.
Alexander Pitchford, Chris Granade, Paul D. Nation & Robert J. Johansson

QuTiP Version:      4.1.0
Numpy Version:      1.12.0
Scipy Version:      0.19.0
Cython Version:     0.25.2
Matplotlib Version: 1.5.2rc2
Python Version:     2.7.13
Number of CPUs:     2
BLAS Info:          OPENBLAS
OPENMP Installed:   True
INTEL MKL Ext:      False
Platform Info:      Linux (x86_64)
Installation path:  /usr/lib64/python2.7/site-packages/qutip-4.1.0-py2.7-linux-x86_64.egg/qutip
在此输入代码...

在 2017年3月14日星期二 UTC+8上午1:36:46，Paul Nation写道：

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+un...@googlegroups.com.

Paul Nation

unread,

Mar 14, 2017, 12:26:09 AM3/14/17

to qu...@googlegroups.com

We do not have any method as of yet for the 4.x releases. The Python monte carlo solver is usually fast enough. But if you really need the Fortran version, then you need to use QuTiP 3.1 for the time being.

lee Albert

March 13, 2017 at 20:16

Thank you, If i really want to install mcf90, would you give me a method? my platform is Fedora 25
QuTiP: Quantum Toolbox in Python Copyright (c) 2011 and later. Alexander Pitchford, Chris Granade, Paul D. Nation & Robert J. Johansson QuTiP Version: 4.1.0 Numpy Version: 1.12.0 Scipy Version: 0.19.0 Cython Version: 0.25.2 Matplotlib Version: 1.5.2rc2 Python Version: 2.7.13 Number of CPUs: 2 BLAS Info: OPENBLAS OPENMP Installed: True INTEL MKL Ext: False Platform Info: Linux (x86_64) Installation path: /usr/lib64/python2.7/site-packages/qutip-4.1.0-py2.7-linux-x86_64.egg/qutip 在此输入代码...

在 2017年3月14日星期二 UTC+8上午1:36:46，Paul Nation写道：

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qutip+un...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Paul Nation

March 13, 2017 at 11:36

It has been removed as it is no longer maintained. It is being moved to its own installer.

-P

lee Albert

March 13, 2017 at 10:45

Did you cancel the --with-f90mc option?
When i compile qutip --with-f90mc error info appeared as `error: option --with-f90mc not recognized ` mcsolver_f90 disappeared.
could you give me more information of 90mc? thanks

在 2017年3月12日星期日 UTC+8上午7:05:20，Paul Nation写道：

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qutip+un...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Paul Nation

March 11, 2017 at 16:05

Alex Pitchford

unread,

Mar 14, 2017, 5:57:14 AM3/14/17

to qu...@googlegroups.com

Dear Lee Albert,

We would be interested to know if/why users would like the fortran MC solver made available again. We were unsure as to how widely it was used. We believed that users had mostly migrated over to the Python version. If you believe that the fortran version has significant advantages to users, then this will provide motivation for prioritising making it available in qutip 4.x

Thank you

Alex

On 14 March 2017 at 04:26, Paul Nation <nonher...@gmail.com> wrote:

We do not have any method as of yet for the 4.x releases. The Python monte carlo solver is usually fast enough. But if you really need the Fortran version, then you need to use QuTiP 3.1 for the time being.

lee Albert

March 13, 2017 at 20:16

Thank you, If i really want to install mcf90, would you give me a method? my platform is Fedora 25
QuTiP: Quantum Toolbox in Python Copyright (c) 2011 and later. Alexander Pitchford, Chris Granade, Paul D. Nation & Robert J. Johansson QuTiP Version: 4.1.0 Numpy Version: 1.12.0 Scipy Version: 0.19.0 Cython Version: 0.25.2 Matplotlib Version: 1.5.2rc2 Python Version: 2.7.13 Number of CPUs: 2 BLAS Info: OPENBLAS OPENMP Installed: True INTEL MKL Ext: False Platform Info: Linux (x86_64) Installation path: /usr/lib64/python2.7/site-packages/qutip-4.1.0-py2.7-linux-x86_64.egg/qutip 在此输入代码...

在 2017年3月14日星期二 UTC+8上午1:36:46，Paul Nation写道：

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Paul Nation

March 13, 2017 at 11:36

It has been removed as it is no longer maintained. It is being moved to its own installer.

-P

lee Albert

March 13, 2017 at 10:45

Did you cancel the --with-f90mc option?
When i compile qutip --with-f90mc error info appeared as `error: option --with-f90mc not recognized ` mcsolver_f90 disappeared.
could you give me more information of 90mc? thanks

在 2017年3月12日星期日 UTC+8上午7:05:20，Paul Nation写道：

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Paul Nation

March 11, 2017 at 16:05

We are pleased to announce that QuTiP 4.1 was released on March 10th 2017. The major highlights are given below:

- MAJOR FEATURE: QuTiP now works for Python 3.5+ on Windows using Visual Studio 2015.

- MAJOR FEATURE: Can now use interpolating cubic splines as time-dependent coefficients.

- MAJOR FEATURE: Sparse matrix - vector multiplication now can run in parallel using OPENMP.

- Automatic tuning of OPENMP threading threashold.

- Cython and other low level code switched to C++ for MS Windows compatibility.

- Partial trace function is now up to 100x+ faster.

- Hermitian verification now up to 100x+ faster.

- Internal Qobj objects now created up to 60x faster.

- Inplace conversion from COO -> CSR sparse formats (e.g. Memory efficiency improvement.)

- Faster reverse Cuthill-Mckee (10x+) and sparse one and inf norms (10x+).

This release also has a few things that break backward compatibility:

- The Fortran monte carlo solver is no longer bundled with QuTiP. It will be moved to its own installation.

- Only Windows installations using Python 3.5+ and Visual Studio 2015 are officially supported.

- The format of the QuTIP configuration file has changed.
--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.

Alex Pitchford

unread,

Mar 14, 2017, 8:47:13 AM3/14/17

to QuTiP: Quantum Toolbox in Python

For those of you who are super keen and testing 4.1 out already:

We are still working on the conda-forge packages. The PyPI (pip) install won't go up until this is complete, as it cannot be changed afterwards, and we may need to make some tweaks for the conda-forge build.

So for now the only install option for 4.1 is to install from source.

The install instructions will be updated when everything works as we would like, and then we will make an update here.

Paul Nation

unread,

Mar 14, 2017, 1:47:39 PM3/14/17

to qu...@googlegroups.com

The Fortran solver is a stop-gap solution. Ideally, we would have a Cython implementation of an ODE solver that does not need to reenter Python at each step in the ODE evolution. An ideal method would be something like Bulirsch-Stoer. Writing ones own ODE solver is usually not a good way to go. However, in this case, if we want faster time-dependent evolution as well, then that is what we need to do. This would be a good project for someone to work on. The only tricky part is making sure the solver can accommodate an arbitrary number of args that could be passed to the RHS function in time-dependent mode.

Alex Pitchford

March 14, 2017 at 03:57

Dear Lee Albert,

We would be interested to know if/why users would like the fortran MC solver made available again. We were unsure as to how widely it was used. We believed that users had mostly migrated over to the Python version. If you believe that the fortran version has significant advantages to users, then this will provide motivation for prioritising making it available in qutip 4.x

Thank you

Alex

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+un...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Paul Nation

March 13, 2017 at 22:26

We do not have any method as of yet for the 4.x releases. The Python monte carlo solver is usually fast enough. But if you really need the Fortran version, then you need to use QuTiP 3.1 for the time being.

lee Albert

March 13, 2017 at 20:16

Thank you, If i really want to install mcf90, would you give me a method? my platform is Fedora 25
QuTiP: Quantum Toolbox in Python Copyright (c) 2011 and later. Alexander Pitchford, Chris Granade, Paul D. Nation & Robert J. Johansson QuTiP Version: 4.1.0 Numpy Version: 1.12.0 Scipy Version: 0.19.0 Cython Version: 0.25.2 Matplotlib Version: 1.5.2rc2 Python Version: 2.7.13 Number of CPUs: 2 BLAS Info: OPENBLAS OPENMP Installed: True INTEL MKL Ext: False Platform Info: Linux (x86_64) Installation path: /usr/lib64/python2.7/site-packages/qutip-4.1.0-py2.7-linux-x86_64.egg/qutip 在此输入代码...

在 2017年3月14日星期二 UTC+8上午1:36:46，Paul Nation写道：
--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+un...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Paul Nation

March 13, 2017 at 11:36

It has been removed as it is no longer maintained. It is being moved to its own installer.

-P

lee Albert

March 13, 2017 at 10:45

Did you cancel the --with-f90mc option?
When i compile qutip --with-f90mc error info appeared as `error: option --with-f90mc not recognized ` mcsolver_f90 disappeared.
could you give me more information of 90mc? thanks

在 2017年3月12日星期日 UTC+8上午7:05:20，Paul Nation写道：
--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+un...@googlegroups.com.

Alex Pitchford

unread,

Mar 17, 2017, 9:38:31 AM3/17/17

to QuTiP: Quantum Toolbox in Python

conda-forge packages are now available. Positively tested on Linux (Python 2.7 & 3.6) and Windows (Python 3.6)

Installing QuTiP should now just be a case of:

conda install qutip

On Saturday, 11 March 2017 23:05:20 UTC, Paul Nation wrote:

Ka Wa Yip

unread,

Mar 17, 2017, 4:55:40 PM3/17/17

to QuTiP: Quantum Toolbox in Python

Hi Alex,

I have qutip installed before via anaconda. I used python 3.4 before. Do I have to delete qutip before reinstalling?

I use Mac Maverick.

Thanks.

Alex Pitchford

unread,

Mar 28, 2017, 6:13:51 PM3/28/17

to QuTiP: Quantum Toolbox in Python

The release is now finished.

All docs should be up-to-date.

The PYPI is updated.

conda-forge packages exist.

On Saturday, 11 March 2017 23:05:20 UTC, Paul Nation wrote:

Wang MinJie

unread,

Apr 23, 2017, 2:20:33 AM4/23/17

to QuTiP: Quantum Toolbox in Python

Dear Paul Nation,

How to update from version4.0.2 to version4.1.0?

在 2017年3月12日星期日 UTC+8上午7:05:20，Paul Nation写道：

We are pleased to announce that QuTiP 4.1 was released on March 10th 2017. The major highlights are given below:

Alex Pitchford

unread,

Apr 24, 2017, 1:29:22 PM4/24/17

to qu...@googlegroups.com

You should just be able to follow the installation instruction for 4.1.0 and it will update automatically.

Alex

--

You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.

souravmajumd...@gmail.com

unread,

May 1, 2017, 10:29:30 AM5/1/17

to QuTiP: Quantum Toolbox in Python

How would i install openmp for qutip

Paul Nation

unread,

May 1, 2017, 10:30:52 AM5/1/17

to QuTiP Group

Install from source.

http://qutip.org/docs/4.1/installation.html

On May 1, 2017 08:29, <souravmajumd...@gmail.com> wrote:

How would i install openmp for qutip

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.

Sourav Majumder

unread,

May 1, 2017, 2:24:09 PM5/1/17

to qu...@googlegroups.com

its saying no module name cython.build

On Mon, May 1, 2017 at 8:00 PM, Paul Nation <nonher...@gmail.com> wrote:

Install from source.

http://qutip.org/docs/4.1/installation.html

On May 1, 2017 08:29, <souravmajumder9432323595@gmail.com> wrote:
How would i install openmp for qutip

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "QuTiP: Quantum Toolbox in Python" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/qutip/LiFH2ZOEZ3Q/unsubscribe.
To unsubscribe from this group and all its topics, send an email to qutip+unsubscribe@googlegroups.com.

Sourav Majumder

unread,

May 1, 2017, 11:56:56 PM5/1/17

to qu...@googlegroups.com

Warning: command line option ‘-Wstrict-prototypes’ is valid for C/ObjC but not for C++

what to do?

On Mon, May 1, 2017 at 8:00 PM, Paul Nation <nonher...@gmail.com> wrote:

Install from source.

http://qutip.org/docs/4.1/installation.html

On May 1, 2017 08:29, <souravmajumder9432323595@gmail.com> wrote:
How would i install openmp for qutip

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to a topic in the Google Groups "QuTiP: Quantum Toolbox in Python" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/qutip/LiFH2ZOEZ3Q/unsubscribe.
To unsubscribe from this group and all its topics, send an email to qutip+unsubscribe@googlegroups.com.

Paul Nation

unread,

May 2, 2017, 12:23:52 AM5/2/17

to QuTiP Group

Update to latest development version on GitHub.

quans...@gmail.com

unread,

May 3, 2017, 3:56:30 AM5/3/17

to QuTiP: Quantum Toolbox in Python

I have installed Anaconda in my PC, QuTiP 4.1 and upgrade NumPy into "1.12.1".But there is some problem when I use the order mcsolve() to solve an example

In [1]: times = np.linspace(0.0, 10.0, 200)

In [2]: psi0 = tensor(fock(2, 0), fock(10, 5))

In [3]: a  = tensor(qeye(2), destroy(10))

In [4]: sm = tensor(destroy(2), qeye(10))

In [5]: H = 2 * np.pi * a.dag() * a + 2 * np.pi * sm.dag() * sm + 2 * np.pi * 0.25 * (sm * a.dag() + sm.dag() * a)

Paul Nation

unread,

May 4, 2017, 12:18:53 AM5/4/17

to qu...@googlegroups.com

The following worked just fine for me using the development 4.2 version and NumPy 1.12.1:

import numpy as np
from qutip import *

times = np.linspace(0, 1, 5)
psi0 = tensor(fock(2, 0), fock(10, 5))
a  = tensor(qeye(2), destroy(10))
sm = tensor(destroy(2), qeye(10))
co = np.sqrt(0.1)*sm
H = 2 * np.pi * a.dag() * a + 2 * np.pi * sm.dag() * sm + \
        2* np.pi * 0.25 * (sm * a.dag() + sm.dag() * a) 
mcsolve(H,psi0,times,c_ops=[co])

quans...@gmail.com

May 3, 2017 at 01:56

In [1]: times = np.linspace(0.0, 10.0, 200)

In [2]: psi0 = tensor(fock(2, 0), fock(10, 5))

In [3]: a = tensor(qeye(2), destroy(10))

In [4]: sm = tensor(destroy(2), qeye(10))

In [5]: H = 2 * np.pi * a.dag() * a + 2 * np.pi * sm.dag() * sm + 2 * np.pi * 0.25 * (sm * a.dag() + sm.dag() * a)

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+un...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Paul Nation

March 11, 2017 at 16:05

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+un...@googlegroups.com.

Quansheng zhung

unread,

May 7, 2017, 4:35:15 AM5/7/17

to qu...@googlegroups.com

The QuTiP toolbox only provide conda install way for windows 7. And when I try to upgrade the toolbox with the following order

conda install -c https://conda.anaconda.org/conda-forge qutip

the old verion 4.1.0 is installed again. So would you supply the whl file for me, which I can use with the pip installation.

By the way, i want to know what is wrong with my python. I have texted my code in other computers, and same error reminder. The following is the way to install the Qutip 4.1.0

1. Install the Anaconda 4.3.1.

2. pip install C:\…\qutip-4.1.0-cp36-cp36m-win.whl

Best wishes.

2017-05-04 12:18 GMT+08:00 Paul Nation <nonher...@gmail.com>:

The following worked just fine for me using the development 4.2 version and NumPy 1.12.1:

import numpy as np
from qutip import *

times = np.linspace(0, 1, 5)
psi0 = tensor(fock(2, 0), fock(10, 5))
a  = tensor(qeye(2), destroy(10))
sm = tensor(destroy(2), qeye(10))
co = np.sqrt(0.1)*sm
H = 2 * np.pi * a.dag() * a + 2 * np.pi * sm.dag() * sm + \
        2* np.pi * 0.25 * (sm * a.dag() + sm.dag() * a) 
mcsolve(H,psi0,times,c_ops=[co])

quans...@gmail.com

May 3, 2017 at 01:56

In [1]: times = np.linspace(0.0, 10.0, 200)

In [2]: psi0 = tensor(fock(2, 0), fock(10, 5))

In [3]: a = tensor(qeye(2), destroy(10))

In [4]: sm = tensor(destroy(2), qeye(10))

In [5]: H = 2 * np.pi * a.dag() * a + 2 * np.pi * sm.dag() * sm + 2 * np.pi * 0.25 * (sm * a.dag() + sm.dag() * a)

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

Paul Nation

March 11, 2017 at 16:05

We are pleased to announce that QuTiP 4.1 was released on March 10th 2017. The major highlights are given below:

- MAJOR FEATURE: QuTiP now works for Python 3.5+ on Windows using Visual Studio 2015.

- MAJOR FEATURE: Can now use interpolating cubic splines as time-dependent coefficients.

- MAJOR FEATURE: Sparse matrix - vector multiplication now can run in parallel using OPENMP.

- Automatic tuning of OPENMP threading threashold.

- Cython and other low level code switched to C++ for MS Windows compatibility.

- Partial trace function is now up to 100x+ faster.

- Hermitian verification now up to 100x+ faster.

- Internal Qobj objects now created up to 60x faster.

- Inplace conversion from COO -> CSR sparse formats (e.g. Memory efficiency improvement.)

- Faster reverse Cuthill-Mckee (10x+) and sparse one and inf norms (10x+).

This release also has a few things that break backward compatibility:

- The Fortran monte carlo solver is no longer bundled with QuTiP. It will be moved to its own installation.

- Only Windows installations using Python 3.5+ and Visual Studio 2015 are officially supported.

- The format of the QuTIP configuration file has changed.

--
You received this message because you are subscribed to the Google Groups "QuTiP: Quantum Toolbox in Python" group.

To unsubscribe from this group and stop receiving emails from it, send an email to qutip+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

You received this message because you are subscribed to a topic in the Google Groups "QuTiP: Quantum Toolbox in Python" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/qutip/LiFH2ZOEZ3Q/unsubscribe.
To unsubscribe from this group and all its topics, send an email to qutip+unsubscribe@googlegroups.com.

Reply all

Reply to author

Forward

Message has been deleted