v3.2.1

[marcopeix]

New features

• [FEAT] Categorical feature support for distributed workflows @marcopeix ([FEAT] Categorical feature support for distributed workflows #1581)
• [FEAT] Support categorical features for explainability @marcopeix ([FEAT] Support categorical features for explainability #1575)
• [FEAT] Support categorical features for simulation paths @marcopeix ([FEAT] Support categorial features for simulation paths #1574)
• [FEAT] Support categorical features for multivariate and recurrent models @marcopeix ([FEAT] Support cat feat for multivariate models #1573)

Bug fixes

• [FIX] Schaake shuffle @elephaint ([FIX] Schaake shuffle #1586)
• [FIX] Initialize static scalers in distributed prediction workers @ShreyanshGoyal (Initialize static scalers in distributed prediction workers #1579)
• [FIX] Support DBFS paths in distributed partition listing @vikas-prasad-cx ([FIX] Support DBFS paths in distributed partition listing #1576)
• [FIX] Set default gpu=1 for Ray @marcopeix ([FIX] Set default gpu=1 for Ray #1572)
• [FIX] Fix loss computation in multi-device training @marcopeix ([FIX] Fix loss computation in multi-device training #1571)
• [FIX] Fix AutoHINT ray serialization issue @marcopeix ([FIX] Fix AutoHINT ray serialization issue #1570)
• [FIX] Use per-series static features in predict_insample @Vedant-Agarwal (fix: use per-series static features in predict_insample #1568)

---

This discussion was created from the release v3.2.1.