A GPU-Accelerated Open-Source Python Package for Calculating Powder Diffraction, Small-Angle-, and Total Scattering with the Debye Scattering Equation - DebyeCalculator v1.0.14

What's New in v1.0.14

• Enhanced Code Refactoring: Introduced compute functions to streamline code organization, reduce redundancies, and improve readability.
• New Functionality: Users can now directly calculate structure functions and pair distribution functions from scattering data, simplifying workflows and enhancing computational efficiency.

Full Changelog: GitHub Compare v1.0.13...v1.0.14

Scientific Software - Peer-reviewed - Jupyter Notebook
Published by FrederikLizakJohansen about 1 year ago

A GPU-Accelerated Open-Source Python Package for Calculating Powder Diffraction, Small-Angle-, and Total Scattering with the Debye Scattering Equation - DebyeCalculator v1.0.13

Changelog for DebyeCalculator v1.0.13

Release Date: 16.09.2024

New Features:

• Support for Double Precision:
  • A new dtype parameter has been added to allow users to specify the precision used in calculations.
  • Double precision (torch.float64) is now supported. This allows for higher accuracy in computations, especially beneficial for calculating small values.
  • The default precision remains as single precision (torch.float32), but users can initialize a DebyeCalculator object with double precision by passing the dtype parameter: python calc = DebyeCalculator(dtype=torch.float64)

Bug Fixes & Stability Improvements:

• Half Precision Instabilities:
  • The introduction of half precision (torch.float16) was tested but resulted in instabilities for small value calculations, so it is not supported in this release.

Scientific Software - Peer-reviewed - Jupyter Notebook
Published by FrederikLizakJohansen over 1 year ago

A GPU-Accelerated Open-Source Python Package for Calculating Powder Diffraction, Small-Angle-, and Total Scattering with the Debye Scattering Equation - DebyeCalculator v1.0.12

Changelog

Support for Pymatgen Structures
DebyeCalculator now supports Pymatgen's Structure objects as input, enabling users to seamlessly work with structures from the Pymatgen library. This includes compatibility for the calculation of I(Q), S(Q), F(Q), and G(r). Users can now pass either ASE atoms objects or Pymatgen structures, improving flexibility in workflows.

Improved Unit Test Coverage
Unit tests have been expanded and enhanced, including better coverage of the partial summation logic for scattering functions and parameter validation for invalid inputs and new input types.

Scientific Software - Peer-reviewed - Jupyter Notebook
Published by FrederikLizakJohansen over 1 year ago

A GPU-Accelerated Open-Source Python Package for Calculating Powder Diffraction, Small-Angle-, and Total Scattering with the Debye Scattering Equation - DebyeCalculator v1.0.11

Changelog

• Tweaks & Fixes: We've fine-tuned performance with some subtle improvements and squashed a few pesky bugs to keep everything running smoothly!

Scientific Software - Peer-reviewed - Jupyter Notebook
Published by FrederikLizakJohansen over 1 year ago

A GPU-Accelerated Open-Source Python Package for Calculating Powder Diffraction, Small-Angle-, and Total Scattering with the Debye Scattering Equation - DebyeCalculator v1.0.10

Changelog

Revamped feature!

• Interactive Mode Reintroduced: The DebyeCalculator interactive interface for Jupyter Notebooks, powered by ipywidgets, is now fully functional again. However, please note that compatibility issues with third-party widgets on Google Colab remain, so this feature is not yet supported in that environment.
• Updated Demo Notebook: The demo notebook has been enhanced to include a new section showcasing the Interactive Mode.

Scientific Software - Peer-reviewed - Jupyter Notebook
Published by FrederikLizakJohansen over 1 year ago

A GPU-Accelerated Open-Source Python Package for Calculating Powder Diffraction, Small-Angle-, and Total Scattering with the Debye Scattering Equation - DebyeCalculator v1.0.9

Changelog

New Features

• Partial Scattering Support: The DebyeCalculator now includes the ability to generate partial scattering from structures. For detailed examples, refer to the Demo.ipynb notebook or the README.
• Demo Notebook: A comprehensive demo covering the major capabilities of DebyeCalculator is now available. You can access it locally by running Demo.ipynb or view it online (see the README for details).

Improvements

• Bug Fixes: Several minor bugs have been resolved.
• Project Configuration: Simplifications and updates have been made to the project.toml file, including renaming for clarity.

Scientific Software - Peer-reviewed - Jupyter Notebook
Published by FrederikLizakJohansen over 1 year ago

A GPU-Accelerated Open-Source Python Package for Calculating Powder Diffraction, Small-Angle-, and Total Scattering with the Debye Scattering Equation - DebyeCalculator 1.0.8

Changelog

Added

• Enhanced form-factors: Updated the form-factor calculations for neutron scattering, improving accuracy (#35).
• Automatic qstep adjustment: Implemented automatic adjustment of the qstep parameter to optimal values when not explicitly provided. Users will receive a UserWarning if the value is considered too large, helping prevent potential issues (3595406).

Fixed

- Dependency Error: Resolved an issue that caused nanoparticle generation to halt due to dependency errors with ASE (#36).

Expect future releases to contain additional functionalities. We are working on implementing some of your great suggestions!

Scientific Software - Peer-reviewed - Jupyter Notebook
Published by FrederikLizakJohansen over 1 year ago

A GPU-Accelerated Open-Source Python Package for Calculating Powder Diffraction, Small-Angle-, and Total Scattering with the Debye Scattering Equation - DebyeCalculator v1.0.7

DebyeCalculator v1.0.7

We are excited to announce the release of DebyeCalculator v1.0.7!

This is the first stable release of DebyeCalculator, a powerful tool for calculating the scattering intensity $I(Q)$ through the Debye scattering equation, the Total Scattering Structure Function $S(Q)$, the Reduced Total Scattering Function $F(Q)$, and the Reduced Atomic Pair Distribution Function $G(r)$ from atomic structures.

This release includes:

• GPU Acceleration: Optimized code for the calculation of the Debye scattering equation on Graphics processing units (GPUs) which accelerate the calculations with orders of magnitudes.
• Support for Multiple Input Formats: Ability to calculate scattering patterns from atomic structures defined in XYZ files, CIF files, or directly as tuples in Python.
• Flexible Parameter Configuration: Easy-to-use interface for updating calculation parameters, including the range and step size for Q values, and the type of radiation used for scattering.
• Documentation: Detailed instructions for installation, usage, and troubleshooting, as well as additional implementation details. Documentations of the latest release can be found here.
• Installation via Pip: DebyeCalculator can be conveniently installed using pip.
• Benchmarking: Test out the package on your local machine or cluster and see the performance you can gain by using DebyeCalculator.

Please refer to the README for more detailed information on how to install and use DebyeCalculator.

If you have any questions or encounter any issues, please don't hesitate to open an issue on our GitHub repository.

Scientific Software - Peer-reviewed - Jupyter Notebook
Published by AndySAnker almost 2 years ago