add LLMs Trainer

chfenglv · Aug 2, 2023 · 9df908d · 9df908d
1 parent 9b586c3
commit 9df908d
Show file tree

Hide file tree

Showing 97 changed files with 7 additions and 4 deletions.
diff --git a/LLM/LLMsTrainer b/LLM/LLMsTrainer
diff --git a/LLM/finetune/assets/ChatGLM Fine-Tune.png → ...glm_finetune/assets/ChatGLM Fine-Tune.png b/LLM/finetune/assets/ChatGLM Fine-Tune.png → ...glm_finetune/assets/ChatGLM Fine-Tune.png
diff --git a/LLM/finetune/assets/label1.png → LLM/chatglm_finetune/assets/label1.png b/LLM/finetune/assets/label1.png → LLM/chatglm_finetune/assets/label1.png
diff --git a/LLM/finetune/assets/label2.png → LLM/chatglm_finetune/assets/label2.png b/LLM/finetune/assets/label2.png → LLM/chatglm_finetune/assets/label2.png
diff --git a/LLM/finetune/assets/playground.png → LLM/chatglm_finetune/assets/playground.png b/LLM/finetune/assets/playground.png → LLM/chatglm_finetune/assets/playground.png
diff --git a/LLM/finetune/data/dataset.jsonl → LLM/chatglm_finetune/data/dataset.jsonl b/LLM/finetune/data/dataset.jsonl → LLM/chatglm_finetune/data/dataset.jsonl
diff --git a/LLM/finetune/data/mixed_dev_dataset.jsonl → ...glm_finetune/data/mixed_dev_dataset.jsonl b/LLM/finetune/data/mixed_dev_dataset.jsonl → ...glm_finetune/data/mixed_dev_dataset.jsonl
diff --git a/LLM/finetune/data/mixed_train_dataset.jsonl → ...m_finetune/data/mixed_train_dataset.jsonl b/LLM/finetune/data/mixed_train_dataset.jsonl → ...m_finetune/data/mixed_train_dataset.jsonl
diff --git a/LLM/finetune/iTrainingLogger.py → LLM/chatglm_finetune/iTrainingLogger.py b/LLM/finetune/iTrainingLogger.py → LLM/chatglm_finetune/iTrainingLogger.py
diff --git a/LLM/finetune/inference.py → LLM/chatglm_finetune/inference.py b/LLM/finetune/inference.py → LLM/chatglm_finetune/inference.py
diff --git a/....github/workflows/build_documentation.yml → ....github/workflows/build_documentation.yml b/....github/workflows/build_documentation.yml → ....github/workflows/build_documentation.yml
diff --git a/...thub/workflows/build_pr_documentation.yml → ...thub/workflows/build_pr_documentation.yml b/...thub/workflows/build_pr_documentation.yml → ...thub/workflows/build_pr_documentation.yml
diff --git a/.../.github/workflows/delete_doc_comment.yml → .../.github/workflows/delete_doc_comment.yml b/.../.github/workflows/delete_doc_comment.yml → .../.github/workflows/delete_doc_comment.yml
diff --git a/.../peft-chatglm/.github/workflows/stale.yml → .../peft-chatglm/.github/workflows/stale.yml b/.../peft-chatglm/.github/workflows/stale.yml → .../peft-chatglm/.github/workflows/stale.yml
diff --git a/.../peft-chatglm/.github/workflows/tests.yml → .../peft-chatglm/.github/workflows/tests.yml b/.../peft-chatglm/.github/workflows/tests.yml → .../peft-chatglm/.github/workflows/tests.yml
diff --git a/LLM/finetune/peft-chatglm/.gitignore → LLM/chatglm_finetune/peft-chatglm/.gitignore b/LLM/finetune/peft-chatglm/.gitignore → LLM/chatglm_finetune/peft-chatglm/.gitignore
diff --git a/LLM/finetune/peft-chatglm/LICENSE → LLM/chatglm_finetune/peft-chatglm/LICENSE b/LLM/finetune/peft-chatglm/LICENSE → LLM/chatglm_finetune/peft-chatglm/LICENSE
diff --git a/LLM/finetune/peft-chatglm/Makefile → LLM/chatglm_finetune/peft-chatglm/Makefile b/LLM/finetune/peft-chatglm/Makefile → LLM/chatglm_finetune/peft-chatglm/Makefile
diff --git a/LLM/finetune/peft-chatglm/README.md → LLM/chatglm_finetune/peft-chatglm/README.md b/LLM/finetune/peft-chatglm/README.md → LLM/chatglm_finetune/peft-chatglm/README.md
diff --git a/LLM/finetune/peft-chatglm/docs/Makefile → ...atglm_finetune/peft-chatglm/docs/Makefile b/LLM/finetune/peft-chatglm/docs/Makefile → ...atglm_finetune/peft-chatglm/docs/Makefile
diff --git a/LLM/finetune/peft-chatglm/docs/README.md → ...tglm_finetune/peft-chatglm/docs/README.md b/LLM/finetune/peft-chatglm/docs/README.md → ...tglm_finetune/peft-chatglm/docs/README.md
diff --git a/...etune/peft-chatglm/docs/source/_config.py → ...etune/peft-chatglm/docs/source/_config.py b/...etune/peft-chatglm/docs/source/_config.py → ...etune/peft-chatglm/docs/source/_config.py
diff --git a/...une/peft-chatglm/docs/source/_toctree.yml → ...une/peft-chatglm/docs/source/_toctree.yml b/...une/peft-chatglm/docs/source/_toctree.yml → ...une/peft-chatglm/docs/source/_toctree.yml
diff --git a/...netune/peft-chatglm/docs/source/index.mdx → ...netune/peft-chatglm/docs/source/index.mdx b/...netune/peft-chatglm/docs/source/index.mdx → ...netune/peft-chatglm/docs/source/index.mdx
diff --git a/...tune/peft-chatglm/docs/source/install.mdx → ...tune/peft-chatglm/docs/source/install.mdx b/...tune/peft-chatglm/docs/source/install.mdx → ...tune/peft-chatglm/docs/source/install.mdx
diff --git a/.../docs/source/package_reference/config.mdx → .../docs/source/package_reference/config.mdx b/.../docs/source/package_reference/config.mdx → .../docs/source/package_reference/config.mdx
diff --git a/...s/source/package_reference/peft_model.mdx → ...s/source/package_reference/peft_model.mdx b/...s/source/package_reference/peft_model.mdx → ...s/source/package_reference/peft_model.mdx
diff --git a/.../docs/source/package_reference/tuners.mdx → .../docs/source/package_reference/tuners.mdx b/.../docs/source/package_reference/tuners.mdx → .../docs/source/package_reference/tuners.mdx
diff --git a/...ne/peft-chatglm/docs/source/quicktour.mdx → ...ne/peft-chatglm/docs/source/quicktour.mdx b/...ne/peft-chatglm/docs/source/quicktour.mdx → ...ne/peft-chatglm/docs/source/quicktour.mdx
diff --git a/...task_guides/image_classification_lora.mdx → ...task_guides/image_classification_lora.mdx b/...task_guides/image_classification_lora.mdx → ...task_guides/image_classification_lora.mdx
diff --git a/...celerate_ds_zero3_cpu_offload_config.yaml → ...celerate_ds_zero3_cpu_offload_config.yaml b/...celerate_ds_zero3_cpu_offload_config.yaml → ...celerate_ds_zero3_cpu_offload_config.yaml
diff --git a/..._clm_accelerate_big_model_inference.ipynb → ..._clm_accelerate_big_model_inference.ipynb b/..._clm_accelerate_big_model_inference.ipynb → ..._clm_accelerate_big_model_inference.ipynb
diff --git a/...t_lora_clm_accelerate_ds_zero3_offload.py → ...t_lora_clm_accelerate_ds_zero3_offload.py b/...t_lora_clm_accelerate_ds_zero3_offload.py → ...t_lora_clm_accelerate_ds_zero3_offload.py
diff --git a/...age_modeling/peft_prefix_tuning_clm.ipynb → ...age_modeling/peft_prefix_tuning_clm.ipynb b/...age_modeling/peft_prefix_tuning_clm.ipynb → ...age_modeling/peft_prefix_tuning_clm.ipynb
diff --git a/...age_modeling/peft_prompt_tuning_clm.ipynb → ...age_modeling/peft_prompt_tuning_clm.ipynb b/...age_modeling/peft_prompt_tuning_clm.ipynb → ...age_modeling/peft_prompt_tuning_clm.ipynb
diff --git a/...causal_language_modeling/requirements.txt → ...causal_language_modeling/requirements.txt b/...causal_language_modeling/requirements.txt → ...causal_language_modeling/requirements.txt
diff --git a/...celerate_ds_zero3_cpu_offload_config.yaml → ...celerate_ds_zero3_cpu_offload_config.yaml b/...celerate_ds_zero3_cpu_offload_config.yaml → ...celerate_ds_zero3_cpu_offload_config.yaml
diff --git a/...tional_generation/peft_adalora_seq2seq.py → ...tional_generation/peft_adalora_seq2seq.py b/...tional_generation/peft_adalora_seq2seq.py → ...tional_generation/peft_adalora_seq2seq.py
diff --git a/...tional_generation/peft_lora_seq2seq.ipynb → ...tional_generation/peft_lora_seq2seq.ipynb b/...tional_generation/peft_lora_seq2seq.ipynb → ...tional_generation/peft_lora_seq2seq.ipynb
diff --git a/...2seq_accelerate_big_model_inference.ipynb → ...2seq_accelerate_big_model_inference.ipynb b/...2seq_accelerate_big_model_inference.ipynb → ...2seq_accelerate_big_model_inference.ipynb
diff --git a/...ra_seq2seq_accelerate_ds_zero3_offload.py → ...ra_seq2seq_accelerate_ds_zero3_offload.py b/...ra_seq2seq_accelerate_ds_zero3_offload.py → ...ra_seq2seq_accelerate_ds_zero3_offload.py
diff --git a/...tion/peft_lora_seq2seq_accelerate_fsdp.py → ...tion/peft_lora_seq2seq_accelerate_fsdp.py b/...tion/peft_lora_seq2seq_accelerate_fsdp.py → ...tion/peft_lora_seq2seq_accelerate_fsdp.py
diff --git a/...neration/peft_prefix_tuning_seq2seq.ipynb → ...neration/peft_prefix_tuning_seq2seq.ipynb b/...neration/peft_prefix_tuning_seq2seq.ipynb → ...neration/peft_prefix_tuning_seq2seq.ipynb
diff --git a/...s/conditional_generation/requirements.txt → ...s/conditional_generation/requirements.txt b/...s/conditional_generation/requirements.txt → ...s/conditional_generation/requirements.txt
diff --git a/...m/examples/image_classification/README.md → ...m/examples/image_classification/README.md b/...m/examples/image_classification/README.md → ...m/examples/image_classification/README.md
diff --git a/...tion/image_classification_peft_lora.ipynb → ...tion/image_classification_peft_lora.ipynb b/...tion/image_classification_peft_lora.ipynb → ...tion/image_classification_peft_lora.ipynb
diff --git a/...ing/Finetune_flan_t5_large_bnb_peft.ipynb → ...ing/Finetune_flan_t5_large_bnb_peft.ipynb b/...ing/Finetune_flan_t5_large_bnb_peft.ipynb → ...ing/Finetune_flan_t5_large_bnb_peft.ipynb
diff --git a/...int8_training/Finetune_opt_bnb_peft.ipynb → ...int8_training/Finetune_opt_bnb_peft.ipynb b/...int8_training/Finetune_opt_bnb_peft.ipynb → ...int8_training/Finetune_opt_bnb_peft.ipynb
diff --git a/...les/int8_training/fine_tune_blip2_int8.py → ...les/int8_training/fine_tune_blip2_int8.py b/...les/int8_training/fine_tune_blip2_int8.py → ...les/int8_training/fine_tune_blip2_int8.py
diff --git a/.../peft_bnb_whisper_large_v2_training.ipynb → .../peft_bnb_whisper_large_v2_training.ipynb b/.../peft_bnb_whisper_large_v2_training.ipynb → .../peft_bnb_whisper_large_v2_training.ipynb
diff --git a/...ples/lora_dreambooth/colab_notebook.ipynb → ...ples/lora_dreambooth/colab_notebook.ipynb b/...ples/lora_dreambooth/colab_notebook.ipynb → ...ples/lora_dreambooth/colab_notebook.ipynb
diff --git a/...reambooth/lora_dreambooth_inference.ipynb → ...reambooth/lora_dreambooth_inference.ipynb b/...reambooth/lora_dreambooth_inference.ipynb → ...reambooth/lora_dreambooth_inference.ipynb
diff --git a/...examples/lora_dreambooth/requirements.txt → ...examples/lora_dreambooth/requirements.txt b/...examples/lora_dreambooth/requirements.txt → ...examples/lora_dreambooth/requirements.txt
diff --git a/...mples/lora_dreambooth/train_dreambooth.py → ...mples/lora_dreambooth/train_dreambooth.py b/...mples/lora_dreambooth/train_dreambooth.py → ...mples/lora_dreambooth/train_dreambooth.py
diff --git a/.../examples/semantic_segmentation/README.md → .../examples/semantic_segmentation/README.md b/.../examples/semantic_segmentation/README.md → .../examples/semantic_segmentation/README.md
diff --git a/...ion/semantic_segmentation_peft_lora.ipynb → ...ion/semantic_segmentation_peft_lora.ipynb b/...ion/semantic_segmentation_peft_lora.ipynb → ...ion/semantic_segmentation_peft_lora.ipynb
diff --git a/...amples/sequence_classification/LoRA.ipynb → ...amples/sequence_classification/LoRA.ipynb b/...amples/sequence_classification/LoRA.ipynb → ...amples/sequence_classification/LoRA.ipynb
diff --git a/...es/sequence_classification/P_Tuning.ipynb → ...es/sequence_classification/P_Tuning.ipynb b/...es/sequence_classification/P_Tuning.ipynb → ...es/sequence_classification/P_Tuning.ipynb
diff --git a/...quence_classification/Prompt_Tuning.ipynb → ...quence_classification/Prompt_Tuning.ipynb b/...quence_classification/Prompt_Tuning.ipynb → ...quence_classification/Prompt_Tuning.ipynb
diff --git a/...quence_classification/prefix_tuning.ipynb → ...quence_classification/prefix_tuning.ipynb b/...quence_classification/prefix_tuning.ipynb → ...quence_classification/prefix_tuning.ipynb
diff --git a/.../sequence_classification/requirements.txt → .../sequence_classification/requirements.txt b/.../sequence_classification/requirements.txt → .../sequence_classification/requirements.txt
diff --git a/..._classification/peft_lora_token_cls.ipynb → ..._classification/peft_lora_token_cls.ipynb b/..._classification/peft_lora_token_cls.ipynb → ..._classification/peft_lora_token_cls.ipynb
diff --git a/...les/token_classification/requirements.txt → ...les/token_classification/requirements.txt b/...les/token_classification/requirements.txt → ...les/token_classification/requirements.txt
diff --git a/LLM/finetune/peft-chatglm/pyproject.toml → ...tglm_finetune/peft-chatglm/pyproject.toml b/LLM/finetune/peft-chatglm/pyproject.toml → ...tglm_finetune/peft-chatglm/pyproject.toml
diff --git a/LLM/finetune/peft-chatglm/scripts/stale.py → ...lm_finetune/peft-chatglm/scripts/stale.py b/LLM/finetune/peft-chatglm/scripts/stale.py → ...lm_finetune/peft-chatglm/scripts/stale.py
diff --git a/LLM/finetune/peft-chatglm/setup.py → LLM/chatglm_finetune/peft-chatglm/setup.py b/LLM/finetune/peft-chatglm/setup.py → LLM/chatglm_finetune/peft-chatglm/setup.py
diff --git a/...inetune/peft-chatglm/src/peft/__init__.py → ...inetune/peft-chatglm/src/peft/__init__.py b/...inetune/peft-chatglm/src/peft/__init__.py → ...inetune/peft-chatglm/src/peft/__init__.py
diff --git a/...une/peft-chatglm/src/peft/import_utils.py → ...une/peft-chatglm/src/peft/import_utils.py b/...une/peft-chatglm/src/peft/import_utils.py → ...une/peft-chatglm/src/peft/import_utils.py
diff --git a/...finetune/peft-chatglm/src/peft/mapping.py → ...finetune/peft-chatglm/src/peft/mapping.py b/...finetune/peft-chatglm/src/peft/mapping.py → ...finetune/peft-chatglm/src/peft/mapping.py
diff --git a/...etune/peft-chatglm/src/peft/peft_model.py → ...etune/peft-chatglm/src/peft/peft_model.py b/...etune/peft-chatglm/src/peft/peft_model.py → ...etune/peft-chatglm/src/peft/peft_model.py
diff --git a/.../peft-chatglm/src/peft/tuners/__init__.py → .../peft-chatglm/src/peft/tuners/__init__.py b/.../peft-chatglm/src/peft/tuners/__init__.py → .../peft-chatglm/src/peft/tuners/__init__.py
diff --git a/...e/peft-chatglm/src/peft/tuners/adalora.py → ...e/peft-chatglm/src/peft/tuners/adalora.py b/...e/peft-chatglm/src/peft/tuners/adalora.py → ...e/peft-chatglm/src/peft/tuners/adalora.py
diff --git a/...tune/peft-chatglm/src/peft/tuners/lora.py → ...tune/peft-chatglm/src/peft/tuners/lora.py b/...tune/peft-chatglm/src/peft/tuners/lora.py → ...tune/peft-chatglm/src/peft/tuners/lora.py
diff --git a/.../peft-chatglm/src/peft/tuners/p_tuning.py → .../peft-chatglm/src/peft/tuners/p_tuning.py b/.../peft-chatglm/src/peft/tuners/p_tuning.py → .../peft-chatglm/src/peft/tuners/p_tuning.py
diff --git a/...-chatglm/src/peft/tuners/prefix_tuning.py → ...-chatglm/src/peft/tuners/prefix_tuning.py b/...-chatglm/src/peft/tuners/prefix_tuning.py → ...-chatglm/src/peft/tuners/prefix_tuning.py
diff --git a/...-chatglm/src/peft/tuners/prompt_tuning.py → ...-chatglm/src/peft/tuners/prompt_tuning.py b/...-chatglm/src/peft/tuners/prompt_tuning.py → ...-chatglm/src/peft/tuners/prompt_tuning.py
diff --git a/...e/peft-chatglm/src/peft/utils/__init__.py → ...e/peft-chatglm/src/peft/utils/__init__.py b/...e/peft-chatglm/src/peft/utils/__init__.py → ...e/peft-chatglm/src/peft/utils/__init__.py
diff --git a/...une/peft-chatglm/src/peft/utils/config.py → ...une/peft-chatglm/src/peft/utils/config.py b/...une/peft-chatglm/src/peft/utils/config.py → ...une/peft-chatglm/src/peft/utils/config.py
diff --git a/...tune/peft-chatglm/src/peft/utils/other.py → ...tune/peft-chatglm/src/peft/utils/other.py b/...tune/peft-chatglm/src/peft/utils/other.py → ...tune/peft-chatglm/src/peft/utils/other.py
diff --git a/...t-chatglm/src/peft/utils/save_and_load.py → ...t-chatglm/src/peft/utils/save_and_load.py b/...t-chatglm/src/peft/utils/save_and_load.py → ...t-chatglm/src/peft/utils/save_and_load.py
diff --git a/LLM/finetune/peft-chatglm/tests/__init__.py → ...m_finetune/peft-chatglm/tests/__init__.py b/LLM/finetune/peft-chatglm/tests/__init__.py → ...m_finetune/peft-chatglm/tests/__init__.py
diff --git a/...une/peft-chatglm/tests/test_common_gpu.py → ...une/peft-chatglm/tests/test_common_gpu.py b/...une/peft-chatglm/tests/test_common_gpu.py → ...une/peft-chatglm/tests/test_common_gpu.py
diff --git a/...inetune/peft-chatglm/tests/test_config.py → ...inetune/peft-chatglm/tests/test_config.py b/...inetune/peft-chatglm/tests/test_config.py → ...inetune/peft-chatglm/tests/test_config.py
diff --git a/...peft-chatglm/tests/test_decoder_models.py → ...peft-chatglm/tests/test_decoder_models.py b/...peft-chatglm/tests/test_decoder_models.py → ...peft-chatglm/tests/test_decoder_models.py
diff --git a/...tglm/tests/test_encoder_decoder_models.py → ...tglm/tests/test_encoder_decoder_models.py b/...tglm/tests/test_encoder_decoder_models.py → ...tglm/tests/test_encoder_decoder_models.py
diff --git a/...e/peft-chatglm/tests/test_gpu_examples.py → ...e/peft-chatglm/tests/test_gpu_examples.py b/...e/peft-chatglm/tests/test_gpu_examples.py → ...e/peft-chatglm/tests/test_gpu_examples.py
diff --git a/...tune/peft-chatglm/tests/testing_common.py → ...tune/peft-chatglm/tests/testing_common.py b/...tune/peft-chatglm/tests/testing_common.py → ...tune/peft-chatglm/tests/testing_common.py
diff --git a/...etune/peft-chatglm/tests/testing_utils.py → ...etune/peft-chatglm/tests/testing_utils.py b/...etune/peft-chatglm/tests/testing_utils.py → ...etune/peft-chatglm/tests/testing_utils.py
diff --git a/LLM/finetune/playground_local.py → LLM/chatglm_finetune/playground_local.py b/LLM/finetune/playground_local.py → LLM/chatglm_finetune/playground_local.py
diff --git a/LLM/finetune/readme.md → LLM/chatglm_finetune/readme.md b/LLM/finetune/readme.md → LLM/chatglm_finetune/readme.md
diff --git a/LLM/finetune/requirements.txt → LLM/chatglm_finetune/requirements.txt b/LLM/finetune/requirements.txt → LLM/chatglm_finetune/requirements.txt
diff --git a/LLM/finetune/train.py → LLM/chatglm_finetune/train.py b/LLM/finetune/train.py → LLM/chatglm_finetune/train.py
diff --git a/LLM/finetune/train.sh → LLM/chatglm_finetune/train.sh b/LLM/finetune/train.sh → LLM/chatglm_finetune/train.sh
diff --git a/LLM/finetune/train_multi_gpu.py → LLM/chatglm_finetune/train_multi_gpu.py b/LLM/finetune/train_multi_gpu.py → LLM/chatglm_finetune/train_multi_gpu.py
diff --git a/LLM/finetune/train_multi_gpu.sh → LLM/chatglm_finetune/train_multi_gpu.sh b/LLM/finetune/train_multi_gpu.sh → LLM/chatglm_finetune/train_multi_gpu.sh
diff --git a/LLM/finetune/utils.py → LLM/chatglm_finetune/utils.py b/LLM/finetune/utils.py → LLM/chatglm_finetune/utils.py
diff --git a/readme.md b/readme.md
@@ -90,24 +90,26 @@
 
 #### 7. 大模型应用（LLM Application）
 
-> 尝试构建大模型（LLM）zero-shot 解决多种任务所需的 prompt pattern(s)。
+> 构建大模型（LLM）zero-shot 解决多种任务所需的 prompt pattern(s)。
 
 | 模型  | 传送门  |
 |---|---|
 | 文本分类（chatglm-6b-Based） | [[这里]](./LLM/zero-shot/readme.md) |
 | 文本匹配（chatglm-6b-Based） | [[这里]](./LLM/zero-shot/readme.md) |
 | 信息抽取（chatglm-6b-Based） | [[这里]](./LLM/zero-shot/readme.md) |
+| 大模型性格测试（LLMs MBTI） | [[这里]](./LLM/llms_mbti/readme.md) |
 
 
 <br>
 
-#### 8. 大模型微调（LLM Finetune）
+#### 8. 大模型训练（LLM Training）
 
-> 对大模型在多个指定任务上进行微调。
+> 大模型训练相关，涵盖预训练，指令微调，奖励模型，强化学习。
 
 | 模型  | 传送门  |
 |---|---|
-| ChatGLM-6B | [[这里]](./LLM/finetune/readme.md) |
+| ChatGLM-6B Finetune | [[这里]](./LLM/chatglm_finetune/readme.md) |
+| 从零开始训练大模型 | [[这里]](./LLM/LLMsTrainer/readme.md) |
 
 
 <br>