Change Log

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

Unreleased

0.1.1 - 2025-08-20

Changed

• Pickled training results go to the Peft output model’s directory.

• Training steps configuration was commented out to allow client projects to keep the HF library defaults.

0.1.0 - 2025-08-06

Removed

• Unsloth configuration and API.

Changed

• Switch build tools to pixi.

• Removed zensols.deeplearn dependency.

• Upgrade dependencies: torch transformers, datasets, accelerate, peft, trl.

• Removed default quantization configuration.

• Add LoRA generator configuration.

• Separate and configure LoRA adapter output directory.

• Add trainer evaluation dataset.

• Disable adding end of sentence token in GenerateTask (turn on by setting train_add_eos is True). Now the SFTTrainer does this by default.

• InstructTask.apply_chat_template defaults to False. Now the SFTTrainer does this by default.

0.0.1 - 2025-05-04

Added

• Initial version.