├── .gitignore ├── .pre-commit-config.yaml ├── CODE_OF_CONDUCT.md ├── LICENSE ├── README.md ├── accelerate_config ├── ddp.yaml ├── infer.yaml ├── zero2.yaml └── zero3.yaml ├── assets ├── InternLM-Xcomposer2-VL-7b_comparison.png ├── LLaVA-Next-Mistral-7b_comparison.png ├── LLaVA-Next-Vicuna-7b_comparison.png ├── Qwen-VL-Chat_comparison.png ├── banner.png ├── draw_radar_chart.ipynb ├── logo.svg └── performance.svg ├── docs ├── CustomizedModel.md ├── EvaluationGuide.md └── TrainingArguments.md ├── pyproject.toml ├── scripts ├── config.sh ├── ddpo_instructblip.sh ├── ddpo_llava.sh ├── ddpo_qwenvl.sh ├── dpo_instructblip.sh ├── dpo_internlmxc2vl7b.sh ├── dpo_llava.sh ├── dpo_llavanext.sh ├── dpo_qwenvl.sh ├── eval │ ├── config.sh │ ├── eval_all.sh │ ├── mathvista.sh │ ├── mmbench.sh │ ├── mmbench_sgl.sh │ ├── mme.sh │ ├── mme_sgl.sh │ ├── mmmu.sh │ ├── mmvet.sh │ ├── mmvet_sgl.sh │ ├── mysql_backup.sh │ ├── pope.sh │ ├── pope_sgl.sh │ ├── seedbench.sh │ ├── seedbench_generate.sh │ ├── seedbench_sgl.sh │ ├── vlrlhf.sql │ └── vqa.sh ├── kto_instructblip.sh ├── kto_llava.sh ├── kto_llavanext.sh ├── kto_qwenvl.sh ├── ppo_qwenvl.sh ├── rm_qwenvl.sh ├── sft_instructblip.sh ├── sft_llava.sh ├── sft_llavanext.sh └── sft_qwenvl.sh └── src └── vlrlhf ├── base ├── __init__.py ├── collator.py ├── model.py ├── processor.py └── trainer.py ├── dpo.py ├── eval ├── mathvista │ ├── calculate.py │ └── eval.py ├── mmbench │ ├── calculate.py │ ├── eval.py │ └── eval_sgl.py ├── mme │ ├── calculate.py │ ├── eval.py │ └── eval_sgl.py ├── mmmu │ ├── calculate.py │ └── eval.py ├── mmvet │ ├── calculate.py │ ├── eval.py │ └── eval_sgl.py ├── pope │ ├── eval.py │ └── eval_sgl.py ├── seedbench │ ├── calculate.py │ ├── eval.py │ ├── eval_generate.py │ ├── eval_generate_sgl.py │ └── extract_choice.py ├── utils.py └── vqa │ └── generate.py ├── merge_peft_model.py ├── models ├── InstructBlip │ └── __init__.py ├── InternLMXC2 │ ├── __init__.py │ ├── build_mlp.py │ ├── configuration_internlm_xcomposer2.py │ ├── modeling_internlm2.py │ ├── modeling_internlm_xcomposer2.py │ ├── tokenization_internlm_xcomposer2.py │ └── zero_to_fp32.py ├── Llava │ └── __init__.py ├── LlavaNext │ └── __init__.py ├── QwenVL │ ├── __init__.py │ ├── configuration_qwen.py │ ├── modeling_qwen.py │ ├── qwen_generation_utils.py │ ├── tokenization_qwen.py │ └── visual.py └── utils.py ├── ppo.py ├── reward_modeling.py ├── sft.py └── utils ├── auto_load.py ├── common.py ├── data.py └── diff_lib.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/.gitignore -------------------------------------------------------------------------------- /.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/.pre-commit-config.yaml -------------------------------------------------------------------------------- /CODE_OF_CONDUCT.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/CODE_OF_CONDUCT.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/LICENSE -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/README.md -------------------------------------------------------------------------------- /accelerate_config/ddp.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/accelerate_config/ddp.yaml -------------------------------------------------------------------------------- /accelerate_config/infer.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/accelerate_config/infer.yaml -------------------------------------------------------------------------------- /accelerate_config/zero2.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/accelerate_config/zero2.yaml -------------------------------------------------------------------------------- /accelerate_config/zero3.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/accelerate_config/zero3.yaml -------------------------------------------------------------------------------- /assets/InternLM-Xcomposer2-VL-7b_comparison.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/assets/InternLM-Xcomposer2-VL-7b_comparison.png -------------------------------------------------------------------------------- /assets/LLaVA-Next-Mistral-7b_comparison.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/assets/LLaVA-Next-Mistral-7b_comparison.png -------------------------------------------------------------------------------- /assets/LLaVA-Next-Vicuna-7b_comparison.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/assets/LLaVA-Next-Vicuna-7b_comparison.png -------------------------------------------------------------------------------- /assets/Qwen-VL-Chat_comparison.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/assets/Qwen-VL-Chat_comparison.png -------------------------------------------------------------------------------- /assets/banner.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/assets/banner.png -------------------------------------------------------------------------------- /assets/draw_radar_chart.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/assets/draw_radar_chart.ipynb -------------------------------------------------------------------------------- /assets/logo.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/assets/logo.svg -------------------------------------------------------------------------------- /assets/performance.svg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/assets/performance.svg -------------------------------------------------------------------------------- /docs/CustomizedModel.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/docs/CustomizedModel.md -------------------------------------------------------------------------------- /docs/EvaluationGuide.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/docs/EvaluationGuide.md -------------------------------------------------------------------------------- /docs/TrainingArguments.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/docs/TrainingArguments.md -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/pyproject.toml -------------------------------------------------------------------------------- /scripts/config.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/config.sh -------------------------------------------------------------------------------- /scripts/ddpo_instructblip.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/ddpo_instructblip.sh -------------------------------------------------------------------------------- /scripts/ddpo_llava.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/ddpo_llava.sh -------------------------------------------------------------------------------- /scripts/ddpo_qwenvl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/ddpo_qwenvl.sh -------------------------------------------------------------------------------- /scripts/dpo_instructblip.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/dpo_instructblip.sh -------------------------------------------------------------------------------- /scripts/dpo_internlmxc2vl7b.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/dpo_internlmxc2vl7b.sh -------------------------------------------------------------------------------- /scripts/dpo_llava.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/dpo_llava.sh -------------------------------------------------------------------------------- /scripts/dpo_llavanext.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/dpo_llavanext.sh -------------------------------------------------------------------------------- /scripts/dpo_qwenvl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/dpo_qwenvl.sh -------------------------------------------------------------------------------- /scripts/eval/config.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/config.sh -------------------------------------------------------------------------------- /scripts/eval/eval_all.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/eval_all.sh -------------------------------------------------------------------------------- /scripts/eval/mathvista.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/mathvista.sh -------------------------------------------------------------------------------- /scripts/eval/mmbench.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/mmbench.sh -------------------------------------------------------------------------------- /scripts/eval/mmbench_sgl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/mmbench_sgl.sh -------------------------------------------------------------------------------- /scripts/eval/mme.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/mme.sh -------------------------------------------------------------------------------- /scripts/eval/mme_sgl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/mme_sgl.sh -------------------------------------------------------------------------------- /scripts/eval/mmmu.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/mmmu.sh -------------------------------------------------------------------------------- /scripts/eval/mmvet.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/mmvet.sh -------------------------------------------------------------------------------- /scripts/eval/mmvet_sgl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/mmvet_sgl.sh -------------------------------------------------------------------------------- /scripts/eval/mysql_backup.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/mysql_backup.sh -------------------------------------------------------------------------------- /scripts/eval/pope.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/pope.sh -------------------------------------------------------------------------------- /scripts/eval/pope_sgl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/pope_sgl.sh -------------------------------------------------------------------------------- /scripts/eval/seedbench.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/seedbench.sh -------------------------------------------------------------------------------- /scripts/eval/seedbench_generate.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/seedbench_generate.sh -------------------------------------------------------------------------------- /scripts/eval/seedbench_sgl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/seedbench_sgl.sh -------------------------------------------------------------------------------- /scripts/eval/vlrlhf.sql: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/vlrlhf.sql -------------------------------------------------------------------------------- /scripts/eval/vqa.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/eval/vqa.sh -------------------------------------------------------------------------------- /scripts/kto_instructblip.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/kto_instructblip.sh -------------------------------------------------------------------------------- /scripts/kto_llava.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/kto_llava.sh -------------------------------------------------------------------------------- /scripts/kto_llavanext.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/kto_llavanext.sh -------------------------------------------------------------------------------- /scripts/kto_qwenvl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/kto_qwenvl.sh -------------------------------------------------------------------------------- /scripts/ppo_qwenvl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/ppo_qwenvl.sh -------------------------------------------------------------------------------- /scripts/rm_qwenvl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/rm_qwenvl.sh -------------------------------------------------------------------------------- /scripts/sft_instructblip.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/sft_instructblip.sh -------------------------------------------------------------------------------- /scripts/sft_llava.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/sft_llava.sh -------------------------------------------------------------------------------- /scripts/sft_llavanext.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/sft_llavanext.sh -------------------------------------------------------------------------------- /scripts/sft_qwenvl.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/scripts/sft_qwenvl.sh -------------------------------------------------------------------------------- /src/vlrlhf/base/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/base/__init__.py -------------------------------------------------------------------------------- /src/vlrlhf/base/collator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/base/collator.py -------------------------------------------------------------------------------- /src/vlrlhf/base/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/base/model.py -------------------------------------------------------------------------------- /src/vlrlhf/base/processor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/base/processor.py -------------------------------------------------------------------------------- /src/vlrlhf/base/trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/base/trainer.py -------------------------------------------------------------------------------- /src/vlrlhf/dpo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/dpo.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mathvista/calculate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mathvista/calculate.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mathvista/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mathvista/eval.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mmbench/calculate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mmbench/calculate.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mmbench/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mmbench/eval.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mmbench/eval_sgl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mmbench/eval_sgl.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mme/calculate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mme/calculate.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mme/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mme/eval.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mme/eval_sgl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mme/eval_sgl.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mmmu/calculate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mmmu/calculate.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mmmu/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mmmu/eval.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mmvet/calculate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mmvet/calculate.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mmvet/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mmvet/eval.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/mmvet/eval_sgl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/mmvet/eval_sgl.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/pope/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/pope/eval.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/pope/eval_sgl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/pope/eval_sgl.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/seedbench/calculate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/seedbench/calculate.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/seedbench/eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/seedbench/eval.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/seedbench/eval_generate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/seedbench/eval_generate.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/seedbench/eval_generate_sgl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/seedbench/eval_generate_sgl.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/seedbench/extract_choice.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/seedbench/extract_choice.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/utils.py -------------------------------------------------------------------------------- /src/vlrlhf/eval/vqa/generate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/eval/vqa/generate.py -------------------------------------------------------------------------------- /src/vlrlhf/merge_peft_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/merge_peft_model.py -------------------------------------------------------------------------------- /src/vlrlhf/models/InstructBlip/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/InstructBlip/__init__.py -------------------------------------------------------------------------------- /src/vlrlhf/models/InternLMXC2/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/InternLMXC2/__init__.py -------------------------------------------------------------------------------- /src/vlrlhf/models/InternLMXC2/build_mlp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/InternLMXC2/build_mlp.py -------------------------------------------------------------------------------- /src/vlrlhf/models/InternLMXC2/configuration_internlm_xcomposer2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/InternLMXC2/configuration_internlm_xcomposer2.py -------------------------------------------------------------------------------- /src/vlrlhf/models/InternLMXC2/modeling_internlm2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/InternLMXC2/modeling_internlm2.py -------------------------------------------------------------------------------- /src/vlrlhf/models/InternLMXC2/modeling_internlm_xcomposer2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/InternLMXC2/modeling_internlm_xcomposer2.py -------------------------------------------------------------------------------- /src/vlrlhf/models/InternLMXC2/tokenization_internlm_xcomposer2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/InternLMXC2/tokenization_internlm_xcomposer2.py -------------------------------------------------------------------------------- /src/vlrlhf/models/InternLMXC2/zero_to_fp32.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/InternLMXC2/zero_to_fp32.py -------------------------------------------------------------------------------- /src/vlrlhf/models/Llava/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/Llava/__init__.py -------------------------------------------------------------------------------- /src/vlrlhf/models/LlavaNext/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/LlavaNext/__init__.py -------------------------------------------------------------------------------- /src/vlrlhf/models/QwenVL/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/QwenVL/__init__.py -------------------------------------------------------------------------------- /src/vlrlhf/models/QwenVL/configuration_qwen.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/QwenVL/configuration_qwen.py -------------------------------------------------------------------------------- /src/vlrlhf/models/QwenVL/modeling_qwen.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/QwenVL/modeling_qwen.py -------------------------------------------------------------------------------- /src/vlrlhf/models/QwenVL/qwen_generation_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/QwenVL/qwen_generation_utils.py -------------------------------------------------------------------------------- /src/vlrlhf/models/QwenVL/tokenization_qwen.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/QwenVL/tokenization_qwen.py -------------------------------------------------------------------------------- /src/vlrlhf/models/QwenVL/visual.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/QwenVL/visual.py -------------------------------------------------------------------------------- /src/vlrlhf/models/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/models/utils.py -------------------------------------------------------------------------------- /src/vlrlhf/ppo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/ppo.py -------------------------------------------------------------------------------- /src/vlrlhf/reward_modeling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/reward_modeling.py -------------------------------------------------------------------------------- /src/vlrlhf/sft.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/sft.py -------------------------------------------------------------------------------- /src/vlrlhf/utils/auto_load.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/utils/auto_load.py -------------------------------------------------------------------------------- /src/vlrlhf/utils/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/utils/common.py -------------------------------------------------------------------------------- /src/vlrlhf/utils/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/utils/data.py -------------------------------------------------------------------------------- /src/vlrlhf/utils/diff_lib.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/TideDra/VL-RLHF/HEAD/src/vlrlhf/utils/diff_lib.py --------------------------------------------------------------------------------