o wi,*@sddlmZddlmZddlZddlZddlm Z m Z ddl m Z ddl mZddlmZddlmZmZmZddlmZmZdd lmZdd lmZdd lmZdd lmZd Z ej!j"e ddej#ej$fddZ%ej!j"e e dddddej&dddddddddddddd dd!d"dd#d$d%e fd&ee'd'e'd(e(d)e(d*eej)d+ee(d,e(d-e*d.e(d/e(d0e(d1e'd2e(d3e+d4e(d5e(d6e(d7e(d8e*dej,f(d9d:Z-d;ej,dej,fd  ? %dDd&ee'd'e'd.e(d/e(d@ee'dAe*dej,fdBdCZ/dS)E)OptionalN)finetunepretrain)MockDataModule) PEFT_STR2CLS)default_finetune_recipe) default_logdefault_resumetensorboard_logger)nemotron_modelnemotron_trainer),distributed_fused_adam_with_cosine_annealing)GarbageCollectionCallback)MegatronCommOverlapCallback)TimingCallbacknemotron4_340bnamereturncCs ttdS)am Factory function to create a Nemotron4 340B model configuration. Returns: run.Config[pl.LightningModule]: Configuration for the Nemotron4 340B model. Examples: CLI usage: $ nemo llm pretrain model=nemotron4_340b ... Python API usage: >>> model_config = model() >>> print(model_config) )version)r NAMErrh/home/ubuntu/sommelier/.venv/lib/python3.10/site-packages/nemo/collections/llm/recipes/nemotron4_340b.pymodel#s r)targetrdefault Tiiz bf16-mixedg? ii iigh㈵>-C6?Fdirrtensor_parallelismpipeline_parallelismpipeline_parallelism_typevirtual_pipeline_parallelismcontext_parallelismsequence_parallelism num_nodesnum_gpus_per_node max_steps precisionaccumulate_grad_batchesgradient_clip_vallimit_test_batcheslimit_val_batcheslog_every_n_stepsval_check_intervalperformance_modecCstj|ttdid|d|d|d|d|d|d|d| d | d | d | d |d |d|d|dttgtjt|||dt||t|ddt | ||||| dt d}|rbt |}|S)a Create a pre-training recipe for Nemotron4 340B model. This function sets up a complete configuration for pre-training, including model, trainer, data, logging, optimization, and resumption settings. Args: dir (Optional[str]): Directory for saving logs and checkpoints. name (str): Name of the pre-training run. tensor_parallelism (int): Degree of tensor model parallelism. pipeline_parallelism (int): Degree of pipeline model parallelism. pipeline_parallelism_type (Optional[torch.dtype]): Data type for pipeline parallelism. virtual_pipeline_parallelism (Optional[int]): Size of virtual pipeline parallelism. context_parallelism (int): Degree of context parallelism. sequence_parallelism (bool): Whether to use sequence parallelism. num_nodes (int): Number of compute nodes to use. num_gpus_per_node (int): Number of GPUs per node. max_steps (int): Maximum number of training steps. precision (str): Precision configuration, one of fp32, 16-mixed or bf16-mixed. accumulate_grad_batches (int): Number of steps per gradient accumulation. gradient_clip_val (float): Value for gradient clipping. limit_test_batches (int): Limit the number of test batches. limit_val_batches (int): Limit the number of validation batches. log_every_n_steps (int): Log every n steps. val_check_interval (int): Run validation every N steps. global_batch_size (int): Global batch size. micro_batch_size (int): Micro batch size. seq_length (int): Sequence length. warmup_steps (int): Number of warmup steps. constant_steps (int): Number of constant steps. min_lr (float): Minimum learning rate. max_lr (float): Maximum learning rate. performance_mode (bool): If true, enables optimizations for maximum performance. fn (Callable): The pre-training function to use. Returns: run.Partial: Partial configuration for pre-training. Examples: CLI usage: $ nemo llm pretrain --factory nemotron4_340b $ nemo llm pretrain --factory "nemotron4_340b(num_nodes=1, name='my_nemotron_pretrain')" Python API usage: >>> recipe = pretrain_recipe(name="nemotron_pretrain", num_nodes=1) >>> print(recipe) Note: This recipe uses a mock dataset, look for the finetune examples to see how to change the dataset. r$r%r&r'r(r)r*r+r,r-r.r0r1r2r3 callbacks) seq_lengthglobal_batch_sizemicro_batch_sizer)r#rr )r- warmup_stepsconstant_stepsmin_lrmax_lr clip_grad)rtrainerdatalogoptimresumeNr) runPartialrr Configrrrr r r "pretrain_performance_optimizations)r#rr$r%r&r'r(r)r*r+r,r-r.r/r0r1r2r3r7r8r6r9r:r;r<r4fnreciperrrpretrain_recipe7spU      'rIrHcCs^|jjsg|j_tjtddd}tjtddddd}|jj||gd|jj_d|j j _ |S)a Create a performance-optimized pre-training recipe for Nemotron4 340B model. This method enables performance optimizations that may not be suitable for all use cases. It builds upon the standard pre-training recipe and adds additional performance enhancements. Args: recipe (run.Partial): Base pre-train recipe to which performance optimizations will be added Returns: run.Partial: Partial configuration for performance-optimized pre-training. Note: Use this method with caution and only when you need maximum performance. It may not be suitable for all hardware configurations or use cases. d)gc_interval_traingc_interval_valTF)tp_comm_overlapdefer_embedding_wgrad_computewgrad_deferral_limit(overlap_param_gather_with_optimizer_step) r>r5rCrErrextendpluginsgrad_reduce_in_fp32rAconfiguse_precision_aware_optimizer)rHgarbage_collection_callbackmcomm_overlap_callbackrrrrFs,  rFlora peft_schemepacked_sequencecCsttd|||||}|dus|dkr+|dksJd|jj_d|jj_d|jj_ n'|dvrKt t ||_ d|jj_d|jj_d |jj_ ntd |d |jj_|S) a Create a fine-tuning recipe for Nemotron4 340B model. This function sets up a complete configuration for fine-tuning, including model, trainer, data, logging, optimization, and resumption settings. The recipe uses LoRA (Low-Rank Adaptation) for efficient fine-tuning, unless peft_scheme is set to None. Args: dir (Optional[str]): Directory for saving logs and checkpoints. name (str): Name of the fine-tuning run. num_nodes (int): Number of compute nodes to use. num_gpus_per_node (int): Number of GPUs per node. peft_scheme (Optional[str]): Name of the peft scheme to use for fine-tuning. Allowed values: 'lora'/'dora'/'none'/None. packed_sequence (Optional[bool]): Packing multiple training sequences into one long sequence for training efficiency. Default sequence length is 2048. Returns: run.Partial: Partial configuration for fine-tuning. Examples: CLI usage: $ nemo llm finetune --factory nemotron3_22b Python API usage: >>> recipe = finetune_recipe(name="nemotron4_340b_finetune", num_nodes=2) >>> print(recipe) Note: This recipe uses the SQuAD dataset for fine-tuning. zmgoin/Nemotron-4-340B-Base-hfNnonerrgh㈵>)rZdorarYr"zUnrecognized peft scheme: F)rrlowerr>strategytensor_model_parallel_sizepipeline_model_parallel_sizerArUlrrCrErpeft ValueErrorcross_entropy_loss_fusion)r#rr*r+r[r\rHrrrfinetune_recipes )         rg)NrrYrrZF)0typingrlightning.pytorchpytorchplnemo_runrCtorchnemo.collections.llm.apirr"nemo.collections.llm.gpt.data.mockrnemo.collections.llm.peftr-nemo.collections.llm.recipes.finetune_defaultr(nemo.collections.llm.recipes.log.defaultrr r %nemo.collections.llm.recipes.nemotronr r 'nemo.collections.llm.recipes.optim.adamr 3nemo.lightning.pytorch.callbacks.garbage_collectionr6nemo.lightning.pytorch.callbacks.megatron_comm_overlaprnemo.utils.exp_managerrrclifactoryrELightningModulerbfloat16strintdtypeboolfloatrDrIrFrgrrrrs               ! /