v2.0.3: bug fix for spectrum settings

This hotfix addresses a bug on spectrum settings that has been observed by Windows users.

Bug fixes

• Fixed the issue where changes to the spectrum settings had no effect.

v2.0.2: "pip install qfit", and play with our new tutorial notebook.

After version 2.0.2, qfit can be installed via pip install qfit! Besides that, we fixes a few bugs and release a new notebook for advanced usages and tips.

Enhancements

• pip install qfit: With upgraded installation profile (pyproject.toml) and automated release pipeline in GitHub, qfit is now on PyPI (Python Package Index)! After this version, qfit can be installed and managed using pip.
• New documentation: A new notebook QFit_Advanced_Tips.ipynb is created to help users to get into the advanced usage of qfit. That includes multi-data file import, crosstalk calibration and a bunch of tips.

Bug fixes

• Fixed the bug that image files can't be loaded.
• Require "qutip < 5.0.0" in the environment

v2.0.1: Bug fixes

We release v2.0.1, aiming for resolving existing issues and enhancing user experiences.

Enhancements

• The size control of the canvas now fits to screens with different DPI and scaling.
• Now the numerical spectrum calculation supports multi-processing and is speeded up.
  • Users can adjust the number of CPUs used for calculation.
  • Added pathos to dependencies to support multi-processing.

Bug fixes

• X and Y axes of the plot are displayed correctly after zoom / pan / reset
• Singular calibration data won't crash the calibration process any more, instead, it will give a warning in the status bar.

QFit v2.0.0: Crosstalk calibration and multi-file fitting

Version 2.0.0, is a major step forward for QFit. This update introduces the crosstalk calibration and fitting against multiple data files in QFit. Moreover, there are a tons of bugs being fixed (and a few new bugs being introduced, such is life).

Enhancements

• New Functionalities
  • Crosstalk Calibration: A superconducting circuit may have multiple tuning knobs such as external magnetic flux threading through circuit loops, and offset charges on metal island biased by gate voltage. QFit helps to map the experimental control parameters (usually voltages) to the circuit parameters (external flux through loops and offset charges) in the Hamiltonian.
  • Multi-File Support: QFit now supports fitting against multiple data files simultaneously. To handle multiple files with great flexibility, we introduce an IMPORT panel, which serves as the first step in the app. In this step, you can import and delete files, view file metadata, select the x and y coordinates and visually check the orientation of the data.
• UI Improvements: Major updates have been made to the user interface, including the implementation of the step IMPORT, to make the experience more intuitive and user-friendly.

Bug Fixes

• Backend Models
  • Calibration Issues: Various bugs related to the calibration process have been addressed, including issues with the calibration button and line edits, to ensure a smoother calibration experience.
• Views
  • UI Fixes: Minor UI issues, such as button enable/disable states and menu visibility, have been corrected to enhance the overall usability of the software.
  • Size of the canvas now gets updated when resizing the app / changing the DPI of the screen. This also includes accurately positioning the axes and the labels, correctly resetting a plot after zoomed-in (clicking home button near the canvas), etc.

Under the Hood

• Measurement Data Model and Control: A new measurement data model has been introduced, along with updates to the measurement data control and view. These changes enable handling of multiple measurement data and fix existing bugs.
• Serialization with dill: Transitioned to dill for serialization to enable saving qfit files with circuit models including custom circuit subsystems.
• Maintain Compatibility: Accept reading saved working progress (.qfit file) from v1.0.x.
• Plot Scaling Improvements: Plot scaling now uses inches as the coordinate system, enhancing the accuracy and readability of graphical data representations.
• Requirements and Dependencies Update: The requirements.txt and setup.py files have been updated to ensure compatibility with the recommended version of Python and other dependencies.
• Enhanced Documentation: The README.md and quick start notebooks have been updated. Some advanced instructions and tips are moved to a separate notebook.

This release focuses on making QFit more robust, reliable, and user-friendly. We acknowledge the introduction of some new bugs with this update and are committed to resolving these in future releases. Your feedback is crucial to us, and we appreciate your support as we continue to enhance QFit to meet your needs. Stay tuned for upcoming updates as we further refine and expand QFit's capabilities.

Release of QFit v1.0.1

We're pleased to announce the release of QFit version 1.0.1. This update introduces a series of improvements and bug fixes that enhance the stability and functionality of QFit, building on the solid foundation set by our initial release.

Enhancements

• Enhanced Documentation: Significant updates to the docstrings for better clarity and developer support, covering views, models, controllers, and utility functions. These improvements aim to facilitate easier navigation and understanding of the codebase for contributors and users alike.

• Functionality Enhancements:

  • In extract, extracted points now automatically snap to the x-grid of the measurement data.
  • In pre-fit, we implemented a "None Selected" option for TRANSITIONS, aiding in the highlighting of sideband transitions and ensuring positive transition frequency is correctly marked.
  • In pre-fit, spectrum calculation is done in a separate thread, enhancing the smoothness of the operation.
  • Now fit process can be interrupted by stop fit buttons
  • Resolved UI issues that affected the responsiveness and user interaction during fitting operations.

Bug fixes

• Debugging and Optimization:
  • Get back the full support for image data, and nearly all of the data processing options applies to images.
  • Addressed several minor bugs and made minor changes to improve the overall robustness and performance of the application.
  • Fixed a display issue for fit-result sweep results and addressed a racing condition in the quantum model's sweepUsage attribute, improving the reliability of data presentation and internal data handling.

Under the hood

• Codebase and Development Improvements:
  • Move all of the models, views and controllers outside of the MainWindow, making the code structure more consistent and easier to maintain.
  • Removed unnecessary files and cleaned up the notebook to streamline the codebase.
  • Various minor naming changes to improve code readability and consistency.

Official release of QFit: v1.0.0

We are pleased to introduce the first official release of QFit, version 1.0.0. Developed with the aim to assist researchers in calibration and parameter fitting for superconducting circuits, QFit simplifies the process of extracting circuit parameters from spectroscopy data. This release follows a period of beta testing, during which we received and incorporated feedback from early users to improve functionality and usability.

Highlights of v1.0.0

• Launch of the First Stable Version: Moving beyond our beta phase, v1.0.0 represents our initial step into providing a stable and reliable tool for the community.

• Streamlined Four-Step Workflow:

  • Calibration: Facilitates the map from voltage inputs into tunable circuit parameters.
  • Point Extraction: Peak extraction and labeling from the spectroscopy data, enhanced with peak snap (Y-Snap), coordinate alignment (X-Snap) and image filters.
  • Interactive Pre-fit: Adjust model parameters with sliders, compare between simulated transitions and experimental data in real-time.
  • Automated Fitting: Refine your parameter fitting with the numerical optimizer, with just a few clicks.

• scQubits Integration: Powered by the scQubits library, QFit supports extensive simulations across various circuit QED systems.

• Improved User Experience: Credit to our UI/UX designer, Tianying Lyu, for crafting an interface that aims to be both intuitive and efficient, catering to researchers' needs.

• Feedback-Driven Development: The insights and suggestions from our community have been crucial in refining QFit. We remain open to and appreciative of continued feedback.

Getting Started

We invite you to explore QFit v1.0.0 and share your experiences or suggestions with us. Your feedback is valuable in guiding future updates and improvements.

For additional information and to download QFit, please visit our GitHub repository. We have prepared you a tutorial notebook QFit_Quick_Start.ipynb for quickly mastering it.

As we make this transition, we continue to encourage and value community feedback to help identify and resolve any lingering issues.