API Reference

This page contains auto-generated API reference documentation [1].

• nemo_rl
  • nemo_rl.data
    • nemo_rl.data.packing
      • nemo_rl.data.packing.algorithms
      • nemo_rl.data.packing.metrics
    • nemo_rl.data.datasets
      • nemo_rl.data.datasets.response_datasets
        • nemo_rl.data.datasets.response_datasets.squad
        • nemo_rl.data.datasets.response_datasets.response_dataset
        • nemo_rl.data.datasets.response_datasets.openmathinstruct2
        • nemo_rl.data.datasets.response_datasets.dapo_math
        • nemo_rl.data.datasets.response_datasets.oasst
        • nemo_rl.data.datasets.response_datasets.gsm8k
        • nemo_rl.data.datasets.response_datasets.oai_format_dataset
        • nemo_rl.data.datasets.response_datasets.deepscaler
        • nemo_rl.data.datasets.response_datasets.clevr
        • nemo_rl.data.datasets.response_datasets.refcoco
        • nemo_rl.data.datasets.response_datasets.nemogym_dataset
        • nemo_rl.data.datasets.response_datasets.geometry3k
        • nemo_rl.data.datasets.response_datasets.tulu3
        • nemo_rl.data.datasets.response_datasets.aime24
        • nemo_rl.data.datasets.response_datasets.helpsteer3
        • nemo_rl.data.datasets.response_datasets.daily_omni
        • nemo_rl.data.datasets.response_datasets.general_conversations_dataset
      • nemo_rl.data.datasets.preference_datasets
        • nemo_rl.data.datasets.preference_datasets.preference_dataset
        • nemo_rl.data.datasets.preference_datasets.binary_preference_dataset
        • nemo_rl.data.datasets.preference_datasets.tulu3
        • nemo_rl.data.datasets.preference_datasets.helpsteer3
      • nemo_rl.data.datasets.eval_datasets
        • nemo_rl.data.datasets.eval_datasets.aime
        • nemo_rl.data.datasets.eval_datasets.mmlu_pro
        • nemo_rl.data.datasets.eval_datasets.gpqa
        • nemo_rl.data.datasets.eval_datasets.local_math_dataset
        • nemo_rl.data.datasets.eval_datasets.mmlu
        • nemo_rl.data.datasets.eval_datasets.math
      • nemo_rl.data.datasets.utils
      • nemo_rl.data.datasets.processed_dataset
      • nemo_rl.data.datasets.raw_dataset
    • nemo_rl.data.dataloader
    • nemo_rl.data.utils
    • nemo_rl.data.interfaces
    • nemo_rl.data.multimodal_utils
    • nemo_rl.data.processors
    • nemo_rl.data.collate_fn
    • nemo_rl.data.chat_templates
    • nemo_rl.data.llm_message_utils
  • nemo_rl.algorithms
    • nemo_rl.algorithms.loss
      • nemo_rl.algorithms.loss.utils
      • nemo_rl.algorithms.loss.loss_functions
      • nemo_rl.algorithms.loss.interfaces
      • nemo_rl.algorithms.loss.wrapper
    • nemo_rl.algorithms.sft
    • nemo_rl.algorithms.utils
    • nemo_rl.algorithms.advantage_estimator
    • nemo_rl.algorithms.grpo
    • nemo_rl.algorithms.distillation
    • nemo_rl.algorithms.async_utils
    • nemo_rl.algorithms.logits_sampling_utils
    • nemo_rl.algorithms.rm
    • nemo_rl.algorithms.reward_functions
    • nemo_rl.algorithms.dpo
  • nemo_rl.models
    • nemo_rl.models.huggingface
      • nemo_rl.models.huggingface.common
    • nemo_rl.models.dtensor
      • nemo_rl.models.dtensor.parallelize
    • nemo_rl.models.megatron
      • nemo_rl.models.megatron.draft
        • nemo_rl.models.megatron.draft.utils
        • nemo_rl.models.megatron.draft.eagle
        • nemo_rl.models.megatron.draft.hidden_capture
      • nemo_rl.models.megatron.setup
      • nemo_rl.models.megatron.data
      • nemo_rl.models.megatron.train
      • nemo_rl.models.megatron.community_import
      • nemo_rl.models.megatron.common
      • nemo_rl.models.megatron.pipeline_parallel
      • nemo_rl.models.megatron.config
    • nemo_rl.models.policy
      • nemo_rl.models.policy.workers
        • nemo_rl.models.policy.workers.base_policy_worker
        • nemo_rl.models.policy.workers.megatron_policy_worker
        • nemo_rl.models.policy.workers.dtensor_policy_worker
        • nemo_rl.models.policy.workers.dtensor_policy_worker_v2
        • nemo_rl.models.policy.workers.patches
      • nemo_rl.models.policy.utils
      • nemo_rl.models.policy.lm_policy
      • nemo_rl.models.policy.interfaces
    • nemo_rl.models.automodel
      • nemo_rl.models.automodel.setup
      • nemo_rl.models.automodel.data
      • nemo_rl.models.automodel.train
      • nemo_rl.models.automodel.config
      • nemo_rl.models.automodel.checkpoint
    • nemo_rl.models.generation
      • nemo_rl.models.generation.vllm
        • nemo_rl.models.generation.vllm.vllm_backend
        • nemo_rl.models.generation.vllm.vllm_worker_async
        • nemo_rl.models.generation.vllm.utils
        • nemo_rl.models.generation.vllm.vllm_worker
        • nemo_rl.models.generation.vllm.vllm_generation
        • nemo_rl.models.generation.vllm.config
      • nemo_rl.models.generation.sglang
        • nemo_rl.models.generation.sglang.sglang_copied_utils
        • nemo_rl.models.generation.sglang.sglang_worker
        • nemo_rl.models.generation.sglang.utils
        • nemo_rl.models.generation.sglang.sglang_generation
        • nemo_rl.models.generation.sglang.config
      • nemo_rl.models.generation.interfaces
  • nemo_rl.distributed
    • nemo_rl.distributed.worker_groups
    • nemo_rl.distributed.virtual_cluster
    • nemo_rl.distributed.ray_actor_environment_registry
    • nemo_rl.distributed.batched_data_dict
    • nemo_rl.distributed.model_utils
    • nemo_rl.distributed.named_sharding
    • nemo_rl.distributed.worker_group_utils
    • nemo_rl.distributed.stateless_process_group
    • nemo_rl.distributed.collectives
  • nemo_rl.evals
    • nemo_rl.evals.eval
    • nemo_rl.evals.answer_parsing
  • nemo_rl.utils
    • nemo_rl.utils.prefetch_venvs
    • nemo_rl.utils.packed_tensor
    • nemo_rl.utils.logger
    • nemo_rl.utils.venvs
    • nemo_rl.utils.nsys
    • nemo_rl.utils.memory_tracker
    • nemo_rl.utils.flops_tracker
    • nemo_rl.utils.flops_formulas
    • nemo_rl.utils.native_checkpoint
    • nemo_rl.utils.nvml
    • nemo_rl.utils.config
    • nemo_rl.utils.timer
    • nemo_rl.utils.checkpoint
  • nemo_rl.experience
    • nemo_rl.experience.rollouts
  • nemo_rl.environments
    • nemo_rl.environments.metrics
    • nemo_rl.environments.utils
    • nemo_rl.environments.code_environment
    • nemo_rl.environments.nemo_gym
    • nemo_rl.environments.code_jaccard_environment
    • nemo_rl.environments.dapo_math_verifier
    • nemo_rl.environments.interfaces
    • nemo_rl.environments.vlm_environment
    • nemo_rl.environments.math_environment
    • nemo_rl.environments.rewards
    • nemo_rl.environments.reward_model_environment
  • nemo_rl.package_info

[1]

Created with sphinx-autodoc2