Yu Xie

# Summary The major changes of this pull request includes 1. Rearrange modules 2. Clean up unused code 3. Merge flare++ into flare 4. Use yaml file for training (same...

Adapted from the Pymatgen PR template. ## Summary Include a summary of major changes in bullet points: * Feature 1 * Feature 2 * Fix 1 * Fix 2 ##...

Adapted from the Pymatgen PR template. ## Summary Include a summary of major changes in bullet points: * Feature 1 * Feature 2 * Fix 1 * Fix 2 ##...

Adapted from the Pymatgen PR template. ## Summary Include a summary of major changes in bullet points: * Feature 1 * Feature 2 * Fix 1 * Fix 2 ##...

## Summary Bug fixes for no stress labels and pylammps sanity check. Pylammps restart feature. B1 descriptors * Fix 1: flare on-the-fly training with no stress label (debug by Julia,...

The mapped variance can suffer from the numerical issues when the variance is very small. Using QR decomposition can significantly improve it. `dot_product.cpp`, `normalized_dot_product.cpp`: remove original code for computing mapped...

Add multiple cutoff support for b1 and b3, previously it was only supported in b2. - `ace_binding.cpp`: add python interface support of b1/b3 multiple cutoffs - `b1.h`, `b1.cpp`, `b3.h`: add...

This pull request wraps up B1, B2, B3 descriptors into a Bk descriptor module. And should be merged after #12 1. When initializing the descriptor, previous setting ```python descriptor_settings =...

# Features Added a new class `ParallelSGP` to build SGP from a large, fixed training data set using MPI, **not on-the-fly construction**. The construction of SGP consists of two processes:...