- users - SasView Mailing Lists

SasView News and a New Release (5.0.6)
by Stephen King - STFC UKRI via users 30 Jun '23

30 Jun '23

Dear Members of the SasView User List, We hope this email finds you safe and well. GENERAL NEWS It is just over a year since our last release of SasView but in that time the Core Developer Team has been quite active. So much so, that when we came to consider a new release we thought long and hard about the best approach: whether to include the proverbial 'kitchen sink' and hope for the best, or to be a little more targeted. In the end we decided to go with the latter and so the new release, SasView 5.0.6, is only a bug fix release. But as you will see below, some of the bugs that have been addressed are ones that had been causing a number of your difficulties. SasView 5.0.6 is still just short of achieving feature-parity with SasView 4.x, but we hope this will happen with the next release. So we would once again like to take the opportunity to encourage and remind any of you still using SasView 4.x to start actively making the transition to SasView 5.x. There is no active development of SasView 4.x, there has not been any for several years, and nor do we now envisage any in the future. In addition, SasView 5.x provides functionality that 4.x does not have. Please note that at the time of writing we are unable to offer a conda package for SasView 5.0.6 due to issues beyond our control, but we are working towards one. In other news: * A very successful Workshop on Magnetic SANS Data Analysis & Software Prospects - obviously with a focus on SasView - was held in Lund in June 2022 (with thanks to Annika Stellhorn for organizing it) which generated a number of excellent suggestions for enhancements to SasView to support this community. Indeed, it led to a follow-up Workshop in April this year. A report on the first Workshop appeared in Neutron News (https://doi.org/10.1080/10448632.2022.2126691) * In October 2022, several of the Core Developer Team attended the first in-person Code Camp held since the Covid-19 pandemic. * And in January 2023 the Core Developer Team was delighted to hear that Paul Butler (NIST NCNR), the driving force behind SasView since its inception, had been awarded a 2022 Silver Medal by the US Department of Commerce for "his global leadership in developing the world's leading software for analyzing small-angle-scattering data" leading to "the first sustainable, community-based scientific software for these techniques". Whilst a significant honour for Paul, this award is also great recognition for the SasView collaboration. NEW RELEASE The Development Team is pleased to announce the release of SasView 5.0.6, available for download as usual through the website at www.sasview.org<http://www.sasview.org>. This release is a bug fix release only but does fix several issues that many Users had found problematic: * The failure of the program to start when installing on a new system due to issues finding the custom_config.py file has been fixed. * The speed with which the program starts up has also been improved. * The bcc_paracrystal and fcc_paracrystal models, which have been labelled as "under review" since 2018, have been checked and corrected and the documentation for all the paracrystal models has been completely reworked. And in this endeavor the SasView Development Team would very much like to express their sincere gratitude to those members of the community that assisted with this task. * Plots now properly support custom data names in the legend. * A problem with the intermittent blanking of plots has also been fixed. * The issue with the LM optimizer failing to run on GPUs or when the starting value of a parameter was outside the min/max range has been fixed. * A number of fitting parameter defaults have been changed to be more realistic. * A number of other issues in the documentation have been corrected and/or updated. NEW MODELS The following models have been added to the Model Marketplace since v5.0.5 was released: * Octahedron_truncated_txtytz * Octahedron_truncated * Rating Curve * Pringle-Schmidt Helices (a revision & documentation update of the existing model) * Lamellar Slab Partition Constant KNOWN ISSUES The rpa (Random Phase Approximation, for polymer scattering) model remains temporarily withdrawn. Of course, as always, there remain many things that could be added, improved or, sadly, that need fixing. So please continue to provide feedback, bug reports (issues), and, when possible, fixes, either to the code or the documentation or tutorials, as GitHub pull requests, for those who can, or by simply emailing help(a)sasview.org<mailto:help@sasview.org>. FUTURE PLANS Another Core Developer Code Camp is being held in September this year, hosted by ISIS/Diamond. And several members of the Core Developer Team will be attending the 25th Anniversary CanSAS Workshop at the ILL/ESRF in October. Meanwhile, plans are currently being developed for a larger Contributor Code Camp (what we are terming Code Camp 'plus') to be held in the US in January 2024. The idea behind this is to facilitate an expansion of the SasView contributor base by providing knowledge and guidance for those who want to be able to help develop the application. The itinerary is still being worked on, but we currently envisage a 7 day workshop that will blend the traditional, task focused development activity of our Core Developer Code Camps with codebase orientations, tutorials, and support activities for new contributors using a hands-on approach. Development activities include tutorial development, documentation development, video development, coding development, website management, and validation and testing activities, as well as many discussions about priorities and future directions for SasView. More information will be posted to this SasView Users list in due course. Thank you for your continued support! And, as ever, if you would like to get involved, please see https://www.sasview.org/contribute/ for further information. The SasView Development Team

1 0

Help needed with testing of the pre-release version of SasView 5.0.6
by Stephen King - STFC UKRI via users 09 Feb '23

09 Feb '23

Dear SasView Users, The Development Team is pleased to announce that the forthcoming new release of SasView, version 5.0.6, is now available for beta-testing. 5.0.6 is strictly a bug fix release addressing many of the top issues identified by the user community, particularly usability issues. At a number of recent conferences, including SAS2022 in Brazil, there has been much discussion about how more of the user base could contribute to this community development effort. Some of the low hanging fruit identified in these discussions were in documentation (verification and editing) and helping to test new releases. In that vein, we are hoping that many of you will be able to find some time to install and test this beta release. Feel free to also ask others you feel would be willing and able to do so. You can download Mac & Windows installers at https://github.com/SasView/sasview/releases/tag/v5.0.6rc1. This 'rc1' stands for 'release candidate 1' but will appear in-program as 'b1' for 'beta 1'. For those of you that are Linux users, we will provide a conda install for the full release but are not doing so for these beta releases for technical reasons. Whilst installing the program, please also take a moment to review the accompanying Release Notes at the link above. The testing window will run until the end of February 2023. Please report any issues when installing/using the program to the usual help(a)sasview.org<mailto:help@sasview.org> or, if you prefer, complete a GitHub bug report at https://github.com/SasView/sasview/issues/new?template=bug_report.md. When reporting back, please make it clear that you are testing version 5.0.6rc1/5.0.6b1 and also provide us with the platform you are using (ie, Windows, Mac) and your specific operating system version. But it would also be great to hear from you, again including the platform/OS you use and what parts of SasView you used, if you do not encounter any problems! Thank you for your continued help and support! The SasView Development Team

1 0

Issue fitting with a GPU in SasView 5.0.5
by Stephen King - STFC UKRI via users 02 Sep '22

02 Sep '22

Dear SasView Users, We have become aware of an issue in the latest SasView 5.0.5 with the new Levenberg-Marquardt (LM) optimiser - the default optimiser - when you are fitting with a GPU enabled. Specifically, if the model you are fitting against contains C code, either embedded in the Python model file or in an ancillary C model file, then nothing happens when you click Fit! The parameters in the FitPage do not update and the fit to the data does not change. You must specifically enable GPU computations, if required, else they are performed on the CPU. So this is a very obvious bug and the chances that you have obtained incorrect results as a consequence are very low. In contrast, if you are only using the CPU (the default condition), or if the model you are fitting is a pure Python model, then this new LM optimiser still works as expected. And if you are using an earlier version of SasView then the old LM optimiser works with a GPU. But if you need to use a GPU for your fitting in SasView 5.0.5 then the workaround is simply to select a different optimiser from Fitting > Fit Algorithms. See https://www.sasview.org/docs/user/qtgui/Perspectives/Fitting/optimizer.html for more information. For the technically-minded, this issue has arisen because we changed the source library for the LM optimiser from SciPy to MPFit. We did this because the SciPy version did not respect parameter bounds during fitting; something which several of you had complained about! Unfortunately, the fact that the MPFit version did not support parallelisation escaped our notice! To tell if your model uses C code, go to https://github.com/SasView/sasmodels/tree/master/sasmodels/models. If there are .py and .c files with the same filename then the model will be affected by this bug. This is, unfortunately, the majority of the library models. The SasView Development Team This email and any attachments are intended solely for the use of the named recipients. If you are not the intended recipient you must not use, disclose, copy or distribute this email or any of its attachments and should notify the sender immediately and delete this email from your system. UK Research and Innovation (UKRI) has taken every reasonable precaution to minimise risk of this email or any attachments containing viruses or malware but the recipient should carry out its own virus and malware checks before opening the attachments. UKRI does not accept any liability for any losses or damages which the recipient may sustain due to presence of any viruses.

1 0

Question on adding structure factors to a custom model
by Mitchell Kennedy via users 15 Jul '22

15 Jul '22

Good Morning. I am examining a system that has a combination of interacting spherical micelles, and a gel network. I am using the gel model and would love to add a spherical model with a squarewell potential structure factor. The only way I have found to do this is to add all three together as separate models, but this is not working very well. My question is, is it possible to add structure factors to custom models or to add a structure factor onto the spherical portion of the model where the radius is calculated using the sphere fitting parameters? Thank you for any help Mitchell Kennedy

2 1

SasView News and a New Release (5.0.5)
by Stephen King - STFC UKRI via users 09 Jun '22

09 Jun '22

Dear Members of the SasView User List, We hope this email finds you safe and well and once again pushing back the frontiers of your science as the restrictions of the pandemic continue to ease. GENERAL NEWS It is almost 6 months since we solicited your help in helping to test the first release candidate (rc1) of SasView 5.0.5, and we are now delighted to say that we have finally released it (see below for further details)! The testing threw up an assortment of issues that, with our limited resources, inevitably took time to rectify, but we have learnt some valuable lessons in the process. Version 5.0.5 is still just short of achieving feature-parity with SasView 4.x, but we are now very close and can see the end of the road. So we would like to take this opportunity to encourage and remind any of you still using SasView 4.x to start actively making the transition to SasView 5.x. There is no active development of SasView 4.x, there has not been any for several years, and nor do we now envisage any in the future. That said, moving to SasView 5.x will have consequences for some of you, particularly those of you using Mac's. In some instances the only way you will be able to run SasView 5.x will be to update your Mac OS or update your Mac. Unfortunately, the need for this is something that is simply beyond our control. There is some good news for those of you that wish to use SasView on Linux. We recently learnt that Maksim Rakitin at the NSLS-II had been producing 64-bit conda packages of SasView for Linux and OSX. If these are of interest, please visit https://github.com/SasView/sasview/wiki/Conda-installation. And we remind you that for some years now, Stuart Prescott & Drew Parson have been maintaining a Debian implementation of SasView. You can find links to this on our Wiki at https://github.com/SasView/sasview/wiki/DevNotes_Projects_Debian. The time spent finalising this new release has naturally impacted progress elsewhere in the collaboration, and so things like the paper on SasView are still not finished. However, there are also some positive developments to report: - We are very pleased to welcome a new developer to the collaboration. Lucas Wilkins is based in the new Research Software Engineering Team at ISIS and will spend up to 80% of his time working on SasView. - Annika Stellhorn at Lund University has put together an in-person Workshop<https://www.linxs.se/events/2022/6/13/linxs-workshop-magnetic-sans-data-ana…> in Lund later this month to explore what additional developments in SasView would be beneficial for the Magnetic-SANS community. This kind of community-led endeavour is something the Development Team is very pleased to see and encourage. - And the Magnetic-SANS Workshop will be followed by a 3-day 'new contributor event<https://indico.esss.lu.se/event/3009/>' for those who want to learn more about the structure of the SasView codebase and how to pursue their own developments. - We have also added a couple of new videos to our YouTube channel<https://www.youtube.com/channel/UCxvD3ysXJ05l6MgY7YKjEFQ> in the past year. Why not take a look! NEW RELEASE The Development Team is pleased to announce the release of SasView 5.0.5, available for download as usual through the website at www.sasview.org<http://www.sasview.org>. This release fixes several issues: * The long standing issue with the Levenberg-Marquardt optimiser not respecting parameter bounds has been resolved by the move to a later version of the Bumps package. * A bug which prevented the radius_effective parameter from being updated in P(Q)*S(Q) models when the data were resolution smeared has been fixed. * A bug that prevented the formation of composite mixture models with multiplicity (for example, models such as core_multi_shell * hardsphere + cylinder or core_multi_shell @ hardsphere + cylinder) has been fixed. * The button to reset the selected Q-limits for fitting now works again! * There has been a technical change to the point in the calculation at which the volume normalisation of P(Q)*S(Q) models is applied by the move to a later version of the Sasmodels package. This change was actually incorporated into v5.0.4 but due to an oversight was omitted from the release notes at the time, although a note was added to the web version after the release. In most instances this change will go un-noticed unless you happen to be plotting the individual contributions of P(Q) or S(Q) and comparing them with similar calculations performed in versions of SasView before v5.0.4, at which point the scaling of the functions will be seen to be different. There are also some new features in this version. Most notably: * The Generic Scattering Calculator Tool has been overhauled and its capabilities significantly expanded. In particular, it will now perform magnetic/polarised SANS computations. As part of this upgrade, coordinate data in some VTK formats are now also supported. * The data loaders have also been improved. SasView will now read CanSAS1D XML data files with multiple <SASdata> blocks in a single <SASentry>. And, by popular demand, 1D data can now be saved in CSV format. SESANS data files with the extension .sesans are now also recognised. * A Boucher-type interfacial profile function has been added to the spherical_sld model. New Models The following models have been added to the Model Marketplace<http://marketplace.sasview.org/> since v5.0.4 was released: * Magnetic Whirl * Maier-Saupe distribution * Cyclic Gaussian distribution * Superball Model - Neither Sphere nor Cube * OrientedMagneticChains Known Issues The 'rpa' (Random Phase Approximation, for polymer scattering) model has been temporarily withdrawn. Some gremlins had crept into how the model parameters interacted with the Fit Page which meant you were not performing the calculation you thought you were. This is being investigated. Should you need it, a separate binary blend model (the most performed calculation) is available from the Model Marketplace<http://marketplace.sasview.org/> . Of course, as always, there remain many things that could be added, improved or, sadly, that need fixing. So please continue to provide feedback, bug reports (issues), and, when possible, fixes, either to the code or the documentation or tutorials, as GitHub pull requests, for those who can, or by simply emailing help(a)sasview.org<mailto:help@sasview.org>. FUTURE PLANS It is likely that some of the Development Team will be present at SAS 2022 this September. If you are also heading there, please do take the opportunity to look us up and say 'Hello'! Thank you for your continued support! The SasView Development Team This email and any attachments are intended solely for the use of the named recipients. If you are not the intended recipient you must not use, disclose, copy or distribute this email or any of its attachments and should notify the sender immediately and delete this email from your system. UK Research and Innovation (UKRI) has taken every reasonable precaution to minimise risk of this email or any attachments containing viruses or malware but the recipient should carry out its own virus and malware checks before opening the attachments. UKRI does not accept any liability for any losses or damages which the recipient may sustain due to presence of any viruses.

1 0

Re: [SasView Users] [SasView Help] help with sasview
by Stephen King - STFC UKRI via users 08 Mar '22

08 Mar '22

Hello Anna, Could you reply to help(a)sasview.org<mailto:help@sasview.org> attaching a failing data file and your SasView log file, please. Also, if you try loading some of the test data shipped with the distribution, does that work? Kind regards, Steve King On behalf of the developers From: help <help-bounces(a)lists.sasview.org> On Behalf Of Anna Stephens via help Sent: 08 March 2022 09:40 To: help(a)sasview.org Subject: [SasView Help] help with sasview Hey, I've downloaded version 5 of sas on my linux machine, when I go to load one of my data .txt files I get the following error in the log explorer: 09:38:12 - ERROR: SasView threw exception: __init__() got an unexpected keyword argument 'nonposx' 09:38:12 - ERROR: SasView threw exception: __init__() got an unexpected keyword argument 'nonposy' I'm not sure how to fix this, can you please help? Thanks so much! Anna This email and any attachments are intended solely for the use of the named recipients. If you are not the intended recipient you must not use, disclose, copy or distribute this email or any of its attachments and should notify the sender immediately and delete this email from your system. UK Research and Innovation (UKRI) has taken every reasonable precaution to minimise risk of this email or any attachments containing viruses or malware but the recipient should carry out its own virus and malware checks before opening the attachments. UKRI does not accept any liability for any losses or damages which the recipient may sustain due to presence of any viruses.

1 0

need help with sasview on linux
by Anna Stephens via users 08 Mar '22

08 Mar '22

Hey, I've downloaded version 5 of sas on my linux machine, when I go to load one of my data .txt files I get the following error in the log explorer: 09:38:12 - ERROR: SasView threw exception: __init__() got an unexpected keyword argument 'nonposx' 09:38:12 - ERROR: SasView threw exception: __init__() got an unexpected keyword argument 'nonposy' I'm not sure how to fix this, can you please help? Thanks so much! Anna

1 0

SasView v5.0.5 - Pre-Release Available for Testing
by Stephen King - STFC UKRI via users 16 Dec '21

16 Dec '21

Dear Colleague, As a developer or trusted user of SasView we would like to elicit your help in road-testing the first release candidate (rc1) of the new version of the program ahead of its full release! You can find MacOS and Windows installers for 5.0.5rc1 here: https://github.com/SasView/sasview/releases/tag/v5.0.5rc1 With this release we are also pleased to welcome six new contributors to the SasView collaboration! Though version 5.0.5 does feature the usual mix of bug fixes, functionality improvements, documentation enhancements, and naturally some new features (more on those in a moment), this version is significant for another reason: it is the first time we will release from GitHub Actions instead of our Travis/Jenkins build system. This change helps future-proof our ability to continue to develop SasView with our limited resources but does come with a one-time penalty. To build SasView with GitHub Actions many of the library packages we use, including the Python version, have had to be upgraded/downgraded. The only way to find out if this has had any undesirable consequences is by putting the program through its paces, and this is where we hope you can assist us! New Features - The ability of the Generic Scattering Calculator Tool to handle magnetic/polarised neutron computations has been completely overhauled and enhanced. As part of this, coordinate data in some VTK formats are now also supported. - The program will now read multiple SASdata blocks in a single SASentry in CanSAS1D XML data files. - In-program data can now also be saved to CSV data files. Functionality Improvements - The data loaders should now be more resilient. - The Q-range 'sliders' are now fully restored, and can be toggled off/on. - The Custom Model Editor now supports some syntax checking and error reporting. - Uncertainty handling on constrained parameters is now propagated correctly. Testing of this release candidate will run from now until mid-January 2022. Please report bugs and comments to help(a)sasview.org<mailto:help@sasview.org>, remembering to specify which platform (Mac/Win) you are using and attaching a copy of your sasview.log file where possible. For any complicated/hard-to-describe bugs, providing the steps to reproduce the issue is extremely helpful. We hope you like SasView 5.0.5, and thank you for your help! The SasView Development Team https://www.sasview.org/ This email and any attachments are intended solely for the use of the named recipients. If you are not the intended recipient you must not use, disclose, copy or distribute this email or any of its attachments and should notify the sender immediately and delete this email from your system. UK Research and Innovation (UKRI) has taken every reasonable precaution to minimise risk of this email or any attachments containing viruses or malware but the recipient should carry out its own virus and malware checks before opening the attachments. UKRI does not accept any liability for any losses or damages which the recipient may sustain due to presence of any viruses.

1 0

Help Needed with Scripting
by Daniel Rosenberg via users 28 Oct '21

28 Oct '21

Hello SASVIEW, I am a researcher at Berkeley Lab at the SIBYLS beamline at the ALS. I am trying to include triaxial fitting from sasmodels into a branch of our pipeline and am having trouble. I was hoping that I might be able to meet with someone in your group to talk it over if someone is available to do so. Thank you for your consideration. Kind Regards Daniel Rosenberg Advanced Light Source SIBYLS Beamline <http://sibyls.als.lbl.gov/> MBIB Division Lawrence Berkeley Lab <http://biosciences.lbl.gov/divisions/mbib/> Biophysics Department UC Berkeley <http://biophysics.berkeley.edu/>

1 0

SasView is hiring!
by Stephen King - STFC UKRI via users 30 Sep '21

30 Sep '21

Thanks to support from the ISIS Pulsed Neutron & Muon Source, we now have a vacancy for a full-time research software engineer to work on SasView. The position will be part of a new Research Software Engineering Team based at ISIS. Full details can be found at: https://www.careersportal.co.uk/UKRI-careers/jobs/research-software-enginee… Interviews are being held on a rolling basis up to the closing date of October 31, 2021. PLEASE NOTE: Interested applicants must apply through the link above. Expressions of interest, resumes, CVs, etc, sent to any of the SasView mailing lists cannot be considered. This email and any attachments are intended solely for the use of the named recipients. If you are not the intended recipient you must not use, disclose, copy or distribute this email or any of its attachments and should notify the sender immediately and delete this email from your system. UK Research and Innovation (UKRI) has taken every reasonable precaution to minimise risk of this email or any attachments containing viruses or malware but the recipient should carry out its own virus and malware checks before opening the attachments. UKRI does not accept any liability for any losses or damages which the recipient may sustain due to presence of any viruses.

1 0