├── .github ├── ISSUE_TEMPLATE │ ├── bug_report.md │ └── feature_request.md ├── PULL_REQUEST_TEMPLATE.md ├── labeler.yml └── workflows │ ├── _run_test.yml │ ├── build-test-publish-wheel.yml │ ├── cherry-pick-release-commit.yml │ ├── cicd-main.yml │ ├── copyright-check.yml │ ├── labeler.yaml │ ├── release-freeze.yml │ ├── release.yaml │ ├── semantic-pull-request.yml │ └── sync_branch.yaml ├── .gitignore ├── .pre-commit-config.yaml ├── CHANGELOG.md ├── CITATION.cff ├── CONTRIBUTING.md ├── Dockerfile ├── LICENSE ├── MANIFEST.in ├── README.md ├── SECURITY.md ├── conftest.py ├── docs ├── README.md ├── RLHFTraining.md ├── assets │ ├── cai_diagram.png │ ├── cai_flow.png │ ├── ppo_arch.png │ └── ppo_diagram.png ├── user-guide-experimental │ └── README.md └── user-guide │ ├── aligner-algo-header.rst │ ├── cai.rst │ ├── dpo.rst │ ├── draftp.rst │ ├── evaluation.rst │ ├── index.rst │ ├── knowledge-distillation.rst │ ├── nemoaligner.rsts │ ├── reinforce.rst │ ├── rlhf.rst │ ├── rs.rst │ ├── sft.rst │ ├── spin.rst │ ├── steerlm.rst │ └── steerlm2.rst ├── examples ├── mm │ ├── clip │ │ ├── conf │ │ │ └── baseline.yaml │ │ ├── test_reward_model.py │ │ └── train_reward_model.py │ └── stable_diffusion │ │ ├── anneal_sdxl.py │ │ ├── conf │ │ ├── draftp_sd.yaml │ │ └── draftp_sdxl.yaml │ │ ├── train_sd_draftp.py │ │ └── train_sdxl_draftp.py └── nlp │ ├── cai │ ├── CritiqueRevisionInstructions.json │ ├── cai_utils.py │ ├── conf │ │ └── inference_service.yaml │ ├── few_shot_samples_example.json │ ├── generate_rl_cai_dataset.py │ ├── generate_sl_cai_dataset.py │ ├── process_anthropic_hh_using_chat_prompt.py │ └── sys_prompt_constitution │ ├── data │ ├── dpo │ │ └── prepare_packed_dpo_dataset.py │ ├── kto │ │ └── preprocess_anthropichh_data.py │ ├── sft │ │ └── remove_long_dialogues.py │ └── steerlm │ │ ├── attribute_annotate.py │ │ ├── common.py │ │ ├── preprocess_helpsteer2_data.py │ │ ├── preprocess_helpsteer_data.py │ │ ├── preprocess_openassistant_data.py │ │ └── process_to_regression_format.py │ ├── gpt │ ├── conf │ │ ├── gpt_dpo.yaml │ │ ├── gpt_knowledge_distillation.yaml │ │ ├── gpt_kto.yaml │ │ ├── gpt_ppo_actor.yaml │ │ ├── gpt_ppo_critic.yaml │ │ ├── gpt_reinforce_actor.yaml │ │ ├── gpt_rs_actor.yaml │ │ ├── gpt_sft.yaml │ │ ├── gpt_spin.yaml │ │ ├── inference_rm.yaml │ │ └── training_rm.yaml │ ├── serve_ppo_critic.py │ ├── serve_reward_model.py │ ├── train_gpt_dpo.py │ ├── train_gpt_knowledge_distillation.py │ ├── train_gpt_kto.py │ ├── train_gpt_ppo_actor.py │ ├── train_gpt_reinforce_actor.py │ ├── train_gpt_rs_actor.py │ ├── train_gpt_sft.py │ ├── train_gpt_spin.py │ ├── train_reward_model.py │ └── train_steerlm2.py │ └── synthetic_data_gen │ ├── compute_topk_logits.py │ └── conf │ └── compute_topk_logits.yaml ├── nemo_aligner ├── __init__.py ├── algorithms │ ├── __init__.py │ ├── critic_server_trainer.py │ ├── dpo.py │ ├── kto.py │ ├── ppo.py │ ├── reinforce.py │ ├── reward_server.py │ ├── rs.py │ ├── spin.py │ └── supervised.py ├── data │ ├── __init__.py │ ├── mm │ │ ├── pickscore_dataset.py │ │ └── text_webdataset.py │ └── nlp │ │ ├── __init__.py │ │ ├── builders.py │ │ ├── datasets.py │ │ ├── scripts │ │ └── undo_special_tokens.py │ │ └── tests │ │ ├── __init__.py │ │ ├── cai_utils_test.py │ │ └── datasets_test.py ├── experimental │ ├── README.md │ └── __init__.py ├── metrics │ ├── __init__.py │ └── common.py ├── models │ ├── __init__.py │ ├── alignable_interface.py │ ├── mm │ │ └── stable_diffusion │ │ │ ├── image_text_rms.py │ │ │ ├── megatron_sd_draftp_model.py │ │ │ └── megatron_sdxl_draftp_model.py │ └── nlp │ │ ├── __init__.py │ │ └── gpt │ │ ├── __init__.py │ │ ├── gpt_reward_model.py │ │ ├── gpt_sft_model.py │ │ ├── gpt_steerlm_model.py │ │ ├── megatron_gpt_critic.py │ │ ├── megatron_gpt_dpo_model.py │ │ ├── megatron_gpt_knowledge_distillation.py │ │ ├── megatron_gpt_kto_model.py │ │ ├── megatron_gpt_ppo_actor.py │ │ ├── megatron_gpt_regression_reward_model.py │ │ ├── megatron_gpt_reinforce_actor.py │ │ ├── megatron_gpt_reward_model.py │ │ ├── megatron_gpt_rs_actor.py │ │ ├── megatron_gpt_spin_model.py │ │ ├── reward_critic_clients.py │ │ └── reward_model_classes.py ├── package_info.py ├── servers │ ├── __init__.py │ ├── constants.py │ └── http_communicator.py ├── testing │ └── utils.py └── utils │ ├── __init__.py │ ├── batch_iterators.py │ ├── distributed.py │ ├── parallel_state.py │ ├── ppo_utils.py │ ├── server_utils.py │ ├── tests │ ├── __init__.py │ ├── distributed_test.py │ ├── ppo_utils_test.py │ ├── text_generation_utils_test.py │ ├── trainer_utils_test.py │ ├── trt_llm_test.py │ └── utils_test.py │ ├── text_generation_utils.py │ ├── train_script_utils.py │ ├── train_utils.py │ ├── trainer_utils.py │ ├── trt_llm.py │ └── utils.py ├── pyproject.toml ├── setup.py ├── setup ├── requirements.txt └── trtllm.patch └── tests ├── __init__.py ├── functional ├── dpo.sh ├── kd.sh ├── ppo.sh ├── reinforce.sh ├── rm.sh ├── run_all.sh ├── sft.sh ├── test_cases │ ├── dpo-llama3 │ ├── dpo-llama3-pack │ ├── kd-llama3 │ ├── ppo-llama3-pp2-reshard │ ├── reinforce-llama3-pp2-reshard │ ├── rm-llama3 │ └── sft-llama3 └── test_data │ ├── dummy-dpo.jsonl │ ├── dummy-sft.jsonl │ ├── dummy_dpo_packed_90.npy │ ├── synthetic-123.jsonl │ ├── synthetic-kd.jsonl │ └── test-rm.jsonl ├── functional_experimental ├── README.md ├── test_cases │ └── .gitkeep └── test_data │ └── .gitkeep ├── run_mpi_unit.sh └── run_unit.sh /.github/ISSUE_TEMPLATE/bug_report.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/ISSUE_TEMPLATE/bug_report.md -------------------------------------------------------------------------------- /.github/ISSUE_TEMPLATE/feature_request.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/ISSUE_TEMPLATE/feature_request.md -------------------------------------------------------------------------------- /.github/PULL_REQUEST_TEMPLATE.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/PULL_REQUEST_TEMPLATE.md -------------------------------------------------------------------------------- /.github/labeler.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/labeler.yml -------------------------------------------------------------------------------- /.github/workflows/_run_test.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/_run_test.yml -------------------------------------------------------------------------------- /.github/workflows/build-test-publish-wheel.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/build-test-publish-wheel.yml -------------------------------------------------------------------------------- /.github/workflows/cherry-pick-release-commit.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/cherry-pick-release-commit.yml -------------------------------------------------------------------------------- /.github/workflows/cicd-main.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/cicd-main.yml -------------------------------------------------------------------------------- /.github/workflows/copyright-check.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/copyright-check.yml -------------------------------------------------------------------------------- /.github/workflows/labeler.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/labeler.yaml -------------------------------------------------------------------------------- /.github/workflows/release-freeze.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/release-freeze.yml -------------------------------------------------------------------------------- /.github/workflows/release.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/release.yaml -------------------------------------------------------------------------------- /.github/workflows/semantic-pull-request.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/semantic-pull-request.yml -------------------------------------------------------------------------------- /.github/workflows/sync_branch.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.github/workflows/sync_branch.yaml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.gitignore -------------------------------------------------------------------------------- /.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/.pre-commit-config.yaml -------------------------------------------------------------------------------- /CHANGELOG.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/CHANGELOG.md -------------------------------------------------------------------------------- /CITATION.cff: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/CITATION.cff -------------------------------------------------------------------------------- /CONTRIBUTING.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/CONTRIBUTING.md -------------------------------------------------------------------------------- /Dockerfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/Dockerfile -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/LICENSE -------------------------------------------------------------------------------- /MANIFEST.in: -------------------------------------------------------------------------------- 1 | include setup/requirements.txt -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/README.md -------------------------------------------------------------------------------- /SECURITY.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/SECURITY.md -------------------------------------------------------------------------------- /conftest.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/conftest.py -------------------------------------------------------------------------------- /docs/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/README.md -------------------------------------------------------------------------------- /docs/RLHFTraining.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/RLHFTraining.md -------------------------------------------------------------------------------- /docs/assets/cai_diagram.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/assets/cai_diagram.png -------------------------------------------------------------------------------- /docs/assets/cai_flow.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/assets/cai_flow.png -------------------------------------------------------------------------------- /docs/assets/ppo_arch.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/assets/ppo_arch.png -------------------------------------------------------------------------------- /docs/assets/ppo_diagram.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/assets/ppo_diagram.png -------------------------------------------------------------------------------- /docs/user-guide-experimental/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide-experimental/README.md -------------------------------------------------------------------------------- /docs/user-guide/aligner-algo-header.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/aligner-algo-header.rst -------------------------------------------------------------------------------- /docs/user-guide/cai.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/cai.rst -------------------------------------------------------------------------------- /docs/user-guide/dpo.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/dpo.rst -------------------------------------------------------------------------------- /docs/user-guide/draftp.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/draftp.rst -------------------------------------------------------------------------------- /docs/user-guide/evaluation.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/evaluation.rst -------------------------------------------------------------------------------- /docs/user-guide/index.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/index.rst -------------------------------------------------------------------------------- /docs/user-guide/knowledge-distillation.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/knowledge-distillation.rst -------------------------------------------------------------------------------- /docs/user-guide/nemoaligner.rsts: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/nemoaligner.rsts -------------------------------------------------------------------------------- /docs/user-guide/reinforce.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/reinforce.rst -------------------------------------------------------------------------------- /docs/user-guide/rlhf.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/rlhf.rst -------------------------------------------------------------------------------- /docs/user-guide/rs.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/rs.rst -------------------------------------------------------------------------------- /docs/user-guide/sft.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/sft.rst -------------------------------------------------------------------------------- /docs/user-guide/spin.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/spin.rst -------------------------------------------------------------------------------- /docs/user-guide/steerlm.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/steerlm.rst -------------------------------------------------------------------------------- /docs/user-guide/steerlm2.rst: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/docs/user-guide/steerlm2.rst -------------------------------------------------------------------------------- /examples/mm/clip/conf/baseline.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/mm/clip/conf/baseline.yaml -------------------------------------------------------------------------------- /examples/mm/clip/test_reward_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/mm/clip/test_reward_model.py -------------------------------------------------------------------------------- /examples/mm/clip/train_reward_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/mm/clip/train_reward_model.py -------------------------------------------------------------------------------- /examples/mm/stable_diffusion/anneal_sdxl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/mm/stable_diffusion/anneal_sdxl.py -------------------------------------------------------------------------------- /examples/mm/stable_diffusion/conf/draftp_sd.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/mm/stable_diffusion/conf/draftp_sd.yaml -------------------------------------------------------------------------------- /examples/mm/stable_diffusion/conf/draftp_sdxl.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/mm/stable_diffusion/conf/draftp_sdxl.yaml -------------------------------------------------------------------------------- /examples/mm/stable_diffusion/train_sd_draftp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/mm/stable_diffusion/train_sd_draftp.py -------------------------------------------------------------------------------- /examples/mm/stable_diffusion/train_sdxl_draftp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/mm/stable_diffusion/train_sdxl_draftp.py -------------------------------------------------------------------------------- /examples/nlp/cai/CritiqueRevisionInstructions.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/cai/CritiqueRevisionInstructions.json -------------------------------------------------------------------------------- /examples/nlp/cai/cai_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/cai/cai_utils.py -------------------------------------------------------------------------------- /examples/nlp/cai/conf/inference_service.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/cai/conf/inference_service.yaml -------------------------------------------------------------------------------- /examples/nlp/cai/few_shot_samples_example.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/cai/few_shot_samples_example.json -------------------------------------------------------------------------------- /examples/nlp/cai/generate_rl_cai_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/cai/generate_rl_cai_dataset.py -------------------------------------------------------------------------------- /examples/nlp/cai/generate_sl_cai_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/cai/generate_sl_cai_dataset.py -------------------------------------------------------------------------------- /examples/nlp/cai/process_anthropic_hh_using_chat_prompt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/cai/process_anthropic_hh_using_chat_prompt.py -------------------------------------------------------------------------------- /examples/nlp/cai/sys_prompt_constitution: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/cai/sys_prompt_constitution -------------------------------------------------------------------------------- /examples/nlp/data/dpo/prepare_packed_dpo_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/data/dpo/prepare_packed_dpo_dataset.py -------------------------------------------------------------------------------- /examples/nlp/data/kto/preprocess_anthropichh_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/data/kto/preprocess_anthropichh_data.py -------------------------------------------------------------------------------- /examples/nlp/data/sft/remove_long_dialogues.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/data/sft/remove_long_dialogues.py -------------------------------------------------------------------------------- /examples/nlp/data/steerlm/attribute_annotate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/data/steerlm/attribute_annotate.py -------------------------------------------------------------------------------- /examples/nlp/data/steerlm/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/data/steerlm/common.py -------------------------------------------------------------------------------- /examples/nlp/data/steerlm/preprocess_helpsteer2_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/data/steerlm/preprocess_helpsteer2_data.py -------------------------------------------------------------------------------- /examples/nlp/data/steerlm/preprocess_helpsteer_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/data/steerlm/preprocess_helpsteer_data.py -------------------------------------------------------------------------------- /examples/nlp/data/steerlm/preprocess_openassistant_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/data/steerlm/preprocess_openassistant_data.py -------------------------------------------------------------------------------- /examples/nlp/data/steerlm/process_to_regression_format.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/data/steerlm/process_to_regression_format.py -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/gpt_dpo.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/gpt_dpo.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/gpt_knowledge_distillation.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/gpt_knowledge_distillation.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/gpt_kto.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/gpt_kto.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/gpt_ppo_actor.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/gpt_ppo_actor.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/gpt_ppo_critic.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/gpt_ppo_critic.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/gpt_reinforce_actor.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/gpt_reinforce_actor.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/gpt_rs_actor.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/gpt_rs_actor.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/gpt_sft.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/gpt_sft.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/gpt_spin.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/gpt_spin.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/inference_rm.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/inference_rm.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/conf/training_rm.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/conf/training_rm.yaml -------------------------------------------------------------------------------- /examples/nlp/gpt/serve_ppo_critic.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/serve_ppo_critic.py -------------------------------------------------------------------------------- /examples/nlp/gpt/serve_reward_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/serve_reward_model.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_gpt_dpo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_gpt_dpo.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_gpt_knowledge_distillation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_gpt_knowledge_distillation.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_gpt_kto.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_gpt_kto.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_gpt_ppo_actor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_gpt_ppo_actor.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_gpt_reinforce_actor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_gpt_reinforce_actor.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_gpt_rs_actor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_gpt_rs_actor.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_gpt_sft.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_gpt_sft.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_gpt_spin.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_gpt_spin.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_reward_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_reward_model.py -------------------------------------------------------------------------------- /examples/nlp/gpt/train_steerlm2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/gpt/train_steerlm2.py -------------------------------------------------------------------------------- /examples/nlp/synthetic_data_gen/compute_topk_logits.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/synthetic_data_gen/compute_topk_logits.py -------------------------------------------------------------------------------- /examples/nlp/synthetic_data_gen/conf/compute_topk_logits.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/examples/nlp/synthetic_data_gen/conf/compute_topk_logits.yaml -------------------------------------------------------------------------------- /nemo_aligner/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/critic_server_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/critic_server_trainer.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/dpo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/dpo.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/kto.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/kto.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/ppo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/ppo.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/reinforce.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/reinforce.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/reward_server.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/reward_server.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/rs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/rs.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/spin.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/spin.py -------------------------------------------------------------------------------- /nemo_aligner/algorithms/supervised.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/algorithms/supervised.py -------------------------------------------------------------------------------- /nemo_aligner/data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/data/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/data/mm/pickscore_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/data/mm/pickscore_dataset.py -------------------------------------------------------------------------------- /nemo_aligner/data/mm/text_webdataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/data/mm/text_webdataset.py -------------------------------------------------------------------------------- /nemo_aligner/data/nlp/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/data/nlp/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/data/nlp/builders.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/data/nlp/builders.py -------------------------------------------------------------------------------- /nemo_aligner/data/nlp/datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/data/nlp/datasets.py -------------------------------------------------------------------------------- /nemo_aligner/data/nlp/scripts/undo_special_tokens.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/data/nlp/scripts/undo_special_tokens.py -------------------------------------------------------------------------------- /nemo_aligner/data/nlp/tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /nemo_aligner/data/nlp/tests/cai_utils_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/data/nlp/tests/cai_utils_test.py -------------------------------------------------------------------------------- /nemo_aligner/data/nlp/tests/datasets_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/data/nlp/tests/datasets_test.py -------------------------------------------------------------------------------- /nemo_aligner/experimental/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/experimental/README.md -------------------------------------------------------------------------------- /nemo_aligner/experimental/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /nemo_aligner/metrics/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/metrics/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/metrics/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/metrics/common.py -------------------------------------------------------------------------------- /nemo_aligner/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/models/alignable_interface.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/alignable_interface.py -------------------------------------------------------------------------------- /nemo_aligner/models/mm/stable_diffusion/image_text_rms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/mm/stable_diffusion/image_text_rms.py -------------------------------------------------------------------------------- /nemo_aligner/models/mm/stable_diffusion/megatron_sd_draftp_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/mm/stable_diffusion/megatron_sd_draftp_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/mm/stable_diffusion/megatron_sdxl_draftp_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/mm/stable_diffusion/megatron_sdxl_draftp_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/gpt_reward_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/gpt_reward_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/gpt_sft_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/gpt_sft_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/gpt_steerlm_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/gpt_steerlm_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_critic.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_critic.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_dpo_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_dpo_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_knowledge_distillation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_knowledge_distillation.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_kto_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_kto_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_ppo_actor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_ppo_actor.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_regression_reward_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_regression_reward_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_reinforce_actor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_reinforce_actor.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_reward_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_reward_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_rs_actor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_rs_actor.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/megatron_gpt_spin_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/megatron_gpt_spin_model.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/reward_critic_clients.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/reward_critic_clients.py -------------------------------------------------------------------------------- /nemo_aligner/models/nlp/gpt/reward_model_classes.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/models/nlp/gpt/reward_model_classes.py -------------------------------------------------------------------------------- /nemo_aligner/package_info.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/package_info.py -------------------------------------------------------------------------------- /nemo_aligner/servers/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/servers/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/servers/constants.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/servers/constants.py -------------------------------------------------------------------------------- /nemo_aligner/servers/http_communicator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/servers/http_communicator.py -------------------------------------------------------------------------------- /nemo_aligner/testing/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/testing/utils.py -------------------------------------------------------------------------------- /nemo_aligner/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/__init__.py -------------------------------------------------------------------------------- /nemo_aligner/utils/batch_iterators.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/batch_iterators.py -------------------------------------------------------------------------------- /nemo_aligner/utils/distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/distributed.py -------------------------------------------------------------------------------- /nemo_aligner/utils/parallel_state.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/parallel_state.py -------------------------------------------------------------------------------- /nemo_aligner/utils/ppo_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/ppo_utils.py -------------------------------------------------------------------------------- /nemo_aligner/utils/server_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/server_utils.py -------------------------------------------------------------------------------- /nemo_aligner/utils/tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /nemo_aligner/utils/tests/distributed_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/tests/distributed_test.py -------------------------------------------------------------------------------- /nemo_aligner/utils/tests/ppo_utils_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/tests/ppo_utils_test.py -------------------------------------------------------------------------------- /nemo_aligner/utils/tests/text_generation_utils_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/tests/text_generation_utils_test.py -------------------------------------------------------------------------------- /nemo_aligner/utils/tests/trainer_utils_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/tests/trainer_utils_test.py -------------------------------------------------------------------------------- /nemo_aligner/utils/tests/trt_llm_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/tests/trt_llm_test.py -------------------------------------------------------------------------------- /nemo_aligner/utils/tests/utils_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/tests/utils_test.py -------------------------------------------------------------------------------- /nemo_aligner/utils/text_generation_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/text_generation_utils.py -------------------------------------------------------------------------------- /nemo_aligner/utils/train_script_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/train_script_utils.py -------------------------------------------------------------------------------- /nemo_aligner/utils/train_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/train_utils.py -------------------------------------------------------------------------------- /nemo_aligner/utils/trainer_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/trainer_utils.py -------------------------------------------------------------------------------- /nemo_aligner/utils/trt_llm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/trt_llm.py -------------------------------------------------------------------------------- /nemo_aligner/utils/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/nemo_aligner/utils/utils.py -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/pyproject.toml -------------------------------------------------------------------------------- /setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/setup.py -------------------------------------------------------------------------------- /setup/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/setup/requirements.txt -------------------------------------------------------------------------------- /setup/trtllm.patch: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/setup/trtllm.patch -------------------------------------------------------------------------------- /tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/functional/dpo.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/dpo.sh -------------------------------------------------------------------------------- /tests/functional/kd.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/kd.sh -------------------------------------------------------------------------------- /tests/functional/ppo.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/ppo.sh -------------------------------------------------------------------------------- /tests/functional/reinforce.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/reinforce.sh -------------------------------------------------------------------------------- /tests/functional/rm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/rm.sh -------------------------------------------------------------------------------- /tests/functional/run_all.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/run_all.sh -------------------------------------------------------------------------------- /tests/functional/sft.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/sft.sh -------------------------------------------------------------------------------- /tests/functional/test_cases/dpo-llama3: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_cases/dpo-llama3 -------------------------------------------------------------------------------- /tests/functional/test_cases/dpo-llama3-pack: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_cases/dpo-llama3-pack -------------------------------------------------------------------------------- /tests/functional/test_cases/kd-llama3: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_cases/kd-llama3 -------------------------------------------------------------------------------- /tests/functional/test_cases/ppo-llama3-pp2-reshard: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_cases/ppo-llama3-pp2-reshard -------------------------------------------------------------------------------- /tests/functional/test_cases/reinforce-llama3-pp2-reshard: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_cases/reinforce-llama3-pp2-reshard -------------------------------------------------------------------------------- /tests/functional/test_cases/rm-llama3: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_cases/rm-llama3 -------------------------------------------------------------------------------- /tests/functional/test_cases/sft-llama3: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_cases/sft-llama3 -------------------------------------------------------------------------------- /tests/functional/test_data/dummy-dpo.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_data/dummy-dpo.jsonl -------------------------------------------------------------------------------- /tests/functional/test_data/dummy-sft.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_data/dummy-sft.jsonl -------------------------------------------------------------------------------- /tests/functional/test_data/dummy_dpo_packed_90.npy: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_data/dummy_dpo_packed_90.npy -------------------------------------------------------------------------------- /tests/functional/test_data/synthetic-123.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_data/synthetic-123.jsonl -------------------------------------------------------------------------------- /tests/functional/test_data/synthetic-kd.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_data/synthetic-kd.jsonl -------------------------------------------------------------------------------- /tests/functional/test_data/test-rm.jsonl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional/test_data/test-rm.jsonl -------------------------------------------------------------------------------- /tests/functional_experimental/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/functional_experimental/README.md -------------------------------------------------------------------------------- /tests/functional_experimental/test_cases/.gitkeep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/functional_experimental/test_data/.gitkeep: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/run_mpi_unit.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/run_mpi_unit.sh -------------------------------------------------------------------------------- /tests/run_unit.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NVIDIA/NeMo-Aligner/HEAD/tests/run_unit.sh --------------------------------------------------------------------------------