ShapeOut 2.20.1

This release fixes a regression that made it impossible to view images from datasets that did not contain mask information.

• fix: regression TypeError when trying to plot contour data

ShapeOut 2.20.0

Thanks to Eoghan O'Connell for implementing phase and amplitude visualization for QPI data!
I also fixed two bugs in the box filtering widget.

• feat: display and handle quantitative phase imaging data
• fix: filter range control not updating limits, displaying value "100" (#183)
• fix: filter range control not updating handles when resized
• enh: improve plotting performance in QuickView polygon selection mode

ShapeOut 2.19.1

In Shape-Out 2.18.0, I claimed to have set the the "buyukurganci-2022" viscosity model as default for Young's modulus estimation. This was not the case and is corrected in this release.

• fix: "buyukurganci-2022" was not actually the default viscosity model
• ref: remove distutils dependency
• docs: add developer's section

ShapeOut 2.19.0

Thanks to @RaghavaAlajangi who implemented the Young's modulus look-up-table selection in the graphical user interface and while doing so also did most of the work upgrading the GUI toolkit from Qt5 to Qt6.
I also added a preferences section for adding S3 object storage access credentials. Previously you had to use a workaround with environment variables to achieve this. I also added the dclab[s3] extras to the install dependencies which adds boto3 (which is now also part of the binary releases).

Detailed list of changes

• feat: support accessing private S3 basins (#174)
• enh: enable lut selection in analysis view plots (#71)
• fix: only allow user to choose "R.exe" in R binary manual search (#173)
• enh: correctly show system R binary in settings
• ref: migrate from pyqt5 to pyqt6 (#165)
• ref: migrate from pkg_resources to importlib.resources (#155)
• setup: support Python 3.12
• setup: new dependency "boto3" via "dclab[s3]""

ShapeOut 2.18.1

This release should bring back support for interfacing with R to perform (G)LMM analyses. Make sure you have R installed and the executable set in the preferences. You might have to add R to your operating system's PATH variable.

• setup: bump dclab to 0.61.3 (replace rpy2 with wrapper) (#73 #161)

ShapeOut 2.18.0

The 2.18 release introduces a breaking change that is relevant for you if you are using the built-in Young's modulus computation (see below). In addition, I fixed a few minor bugs that partially affect performance when accessing remote datasets.

• BREAKING CHANGE: set default viscosity model for Young's modulus
  computation to "buyukurganci-2022", because it is more accurate
  and also because it is defined for the "0.83% MC-PBS" medium
• BREAKING CHANGE: remove support for machine learning extensions
  (.modc files), because they are not used
• fix: remove leading/trailing slashes for DCOR servers
• setup: bump dclab to 0.61.1 (fix basin-based array access, export from
  datasets with internal basins, other minor fixes)

ShapeOut 2.17.0

With dclab 0.60.0, Shape-Out now supports internal basin features and writes logs to exported data. Make sure to update Shape-Out if you intend to work with the next releases of dcnum or ChipStream.

ShapeOut 2.16.5

Install this release if you are working with basin features (e.g. DCOR data). I made several fixes in dclab 0.59.1 which were related to cyclic basin definitions and data export from basin-based datasets.

ShapeOut 2.16.4

This releases improves basin-compatibility with the dclab 0.59.0 upgrade which I highly recommend. This update is mandatory if you are working with ChipStream 0.4.0 or above.

ShapeOut 2.16.3

This release fixes a few minor bugs and improves usability. With dclab version 0.58.7, you can also expect minor performance improvements in some edge cases.

The builds for macOS and Windows that you can download from this page now have the statistical significance testing capabilities stripped from them. If you would like to use those, you would have to install Shape-Out via Python/pip. For progress on this issue, please watch #161.

• fix: data export feature selection must default to features_loaded
• enh: only show local features in QuickView Event Feature tab
• enh: show dcnum pipeline identifiers in Analysis view
• enh: add section:key tool tips for analysis view metadata
• enh: added links to docs explaining default export selection
• docs: remove Shape-Out 1 reference
• setup: remove rpy2 from dependencies until #161 is fixed
• setup: migrate from setup.py to pyproject.toml
• setup: bump dclab to 0.58.7 (performance)