Castl

Features 🚀
- Implemented radial velocity fitting
- Added functionality to save best-fit parameters
- Archived on Zenodo for better accessibility
- Designed and integrated a new logo

Bug Fixes & Improvements 🛠
- Reduced excessive significant figures in corner plots for improved readability

Thank you for your support. We hope these enhancements improve your experience with castl. Please feel free to reach out to us if you have any questions or encounter any issues.

- The castl Development Team

**Features** 🚀
- **Efficient Interpolation**: Added `LinearNDInterpolator` for models with fewer than 5 parameters.
- **Faster Model Loading**: Introduced `h5grid`, a tool for preloading models.
- **Optimized MCMC Moves**: Removed scaling for `StretchMove` and replaced it with gamma stepping for better convergence.
- **New Parameter**: Added `log(R²/D²)` as an additional model parameter.
- **Flux Unit Conversion**: Automatically converts observed flux units to match model flux units.
- **Smarter Initial Prior**: Uses the best grid point as the starting prior for MCMC sampling.
- **Refined Interpolation Grid**: Reduces grid size around the best grid point for better accuracy.
- **Adjustable Grid Resolution**: Allows setting the number of grid points in each dimension.
- **Enhanced Output**: Saves the best-fit interpolated spectra for future analysis.
- **Expanded File Support**: Now supports additional file types.

**Bug Fixes & Improvements** 🛠
- **Reduced RAM Usage**: Optimized memory handling for better performance.
- **Improved Visualization**: Now saves separate figures for spectra and corner plots.
- **Better Documentation**: Provided a more in-depth example for users.
- **Unit Handling**: Added unit variables for both wavelength and flux.
- **Autocorrelation Fix**: Removed autocorrelation issues in sampling.
- **Code Enhancements**: Improved comments for better readability and maintainability.

Thank you for your support. We hope these enhancements improve your experience with castl. Please feel free to reach out to us if you have any questions or encounter any issues.

_- The castl Development Team_

We are pleased to announce the initial beta release of castl (Computational Analysis of Spectral TempLates) version 0.5.0. Here are the key features in this version:

**Features:**
- The entire package is integrated into a single command.
- Uploaded to PyPI (pip) for easy installation.
- Utilizes the RBFInterpolator for efficient interpolation.
- Monitoring functionality has been added as a feature.
- Optimized for high performance with minimal computational needs.
- Supports multiprocessing for multi-object fits.
- Example code is provided for easy usage.
- Statistical calculations now include uncertainties.
- The grid is automatically normalized.
- Smoothing of model spectra is an additional parameter.
- Auto-correlation has been included.

In this release, we have tested castl with over 50 observed spectra on multiple models. We are committed to continuously improving castl, and our team will continue testing and improving performance.

Thank you for your support. We hope these enhancements improve your experience with castl. Please feel free to reach out to us if you have any questions or encounter any issues.

- The castl Development Team