Flexeval

What's Changed
* Add `developer_message` to OpenAI API Language Model by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/156
* Refactoring tests for `LanguageModel` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/157
* Fix bugs on `evaluate_chat_response` with OpenAI APIs by junya-takayama in https://github.com/sbintuitions/flexeval/pull/155
* Change `LanguageModel._batch_*` methods to private by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/158
* Fix a test function signature in `test_optnai_batch_api.py` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/159
* Add comments to code by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/162
* Add `StringProcessor` to `LanguageModel` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/163


**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.10.2...v0.10.3

What's Changed
* Sarashina2.2 の評価追加 by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/151
* Compute category-wise scores in `ExactMatch` and `SubstringMatch` if specified by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/153
* Handle category with list[str] in LLMScore by kevin3314 in https://github.com/sbintuitions/flexeval/pull/154


**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.10.1...v0.10.2

What's Changed
* Avoid using MPS for pre MacOS_14_0 by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/145
* Introduce `decode_for_lm_continuation` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/146
* Add tokenizers by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/148
* Update `SubstringMatch` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/147
* Update the directory structure for `StringProcessor` and `Tokenizer` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/149
* Stop at BOS in `HuggingFaceLM` and `VLLM` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/150


**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.10.0...v0.10.1

What's Changed
* Add `load_only_first_n` in `ChatbotBench` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/138
* Implement compute_log_probs of OpenAIChatAPI and OpenAIChatBatchAPI by m-ast in https://github.com/sbintuitions/flexeval/pull/131
* Execute OpenAI-related tests on GitHub CI by junya-takayama in https://github.com/sbintuitions/flexeval/pull/139
* Fix some bugs related to `OpenAIChatAPI` and `OpenAIChatBatchAPI` by junya-takayama in https://github.com/sbintuitions/flexeval/pull/141
* Support Claude, Gemini and etc. via `LiteLLM` by junya-takayama in https://github.com/sbintuitions/flexeval/pull/127
* Use `LMOutput` for the output of `LanguageModel` by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/140
* Move `string_processor` under the `core` directory by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/144
* Save `finish_reason` statistics by ryokan0123 in https://github.com/sbintuitions/flexeval/pull/143


**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.9.4...v0.10.0

What's Changed
* Upgrade vLLM to 0.7.2 by butsugiri in https://github.com/sbintuitions/flexeval/pull/132
* fix OpenAIChatBatchAPI class by losyer in https://github.com/sbintuitions/flexeval/pull/134
* fix template for flexeval_reward test by losyer in https://github.com/sbintuitions/flexeval/pull/133
* re-fix OpenAIChatBatchAPI class by losyer in https://github.com/sbintuitions/flexeval/pull/136
* Use sudachipy v0.6.10 or later to fix FTBFS in aarch64 Linux environment by takaswie in https://github.com/sbintuitions/flexeval/pull/137

New Contributors
* losyer made their first contribution in https://github.com/sbintuitions/flexeval/pull/134
* takaswie made their first contribution in https://github.com/sbintuitions/flexeval/pull/137

**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.9.3...v0.9.4

What's Changed
* Fix `get_most_repeated_pattern` by butsugiri in https://github.com/sbintuitions/flexeval/pull/126
* Fix and add tests for batch_compute_chat_log_probs of HuggingFaceLM and VLLM by m-ast in https://github.com/sbintuitions/flexeval/pull/129
* Add new metric functions: LLMGEvalScore and ChatLLMGEvalScore by m-ast in https://github.com/sbintuitions/flexeval/pull/125
* TemplateRendererの追加 by Ktakuya332C in https://github.com/sbintuitions/flexeval/pull/130

New Contributors
* m-ast made their first contribution in https://github.com/sbintuitions/flexeval/pull/129

**Full Changelog**: https://github.com/sbintuitions/flexeval/compare/v0.9.2...v0.9.3