Hugging Face's logo

Transformers documentation

맘바

Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

맘바

개요

맘바(Mamba) 모델은 Albert Gu, Tri Dao가 제안한 맘바: 선택적 상태 공간을 이용한 선형 시간 시퀀스 모델링라는 논문에서 소개 되었습니다.

이 모델은 state-space-models을 기반으로 한 새로운 패러다임 아키텍처입니다. 직관적인 이해를 얻고 싶다면 이곳을 참고 하세요.

해당 논문의 초록입니다:

현재 딥러닝에서 흥미로운 응용 프로그램을 구동하는 대부분의 기초 모델들은 거의 보편적으로 트랜스포머 아키텍처와 그 핵심 어텐션 모듈을 기반으로 합니다. 선형 어텐션, 게이트된 컨볼루션과 순환 모델, 구조화된 상태 공간 모델(SSM) 등 많은 준이차시간(subquadratic-time) 아키텍처가 긴 시퀀스에 대한 트랜스포머의 계산 비효율성을 해결하기 위해 개발되었지만, 언어와 같은 중요한 양식에서는 어텐션만큼 성능을 내지 못했습니다. 우리는 이러한 모델의 주요 약점이 내용 기반 추론을 수행하지 못한다는 점임을 알고 몇 가지를 개선했습니다. 첫째, SSM 매개변수를 입력의 함수로 만드는 것만으로도 이산 모달리티(discrete modalities)의 약점을 해결할 수 있어, 현재 토큰에 따라 시퀀스 길이 차원을 따라 정보를 선택적으로 전파하거나 잊을 수 있게 합니다. 둘째, 이러한 변경으로 효율적인 컨볼루션을 사용할 수 없게 되었지만, 우리는 순환 모드에서 하드웨어를 인식하는 병렬 알고리즘을 설계했습니다. 우리는 이러한 선택적 SSM을 어텐션이나 MLP 블록도 없는 단순화된 종단간 신경망 아키텍처인 맘바에 통합시켰습니다. 맘바는 빠른 추론(트랜스포머보다 5배 높은 처리량)과 시퀀스 길이에 대한 선형 확장성을 누리며, 백만 길이 시퀀스까지 실제 데이터에서 성능이 향상됩니다. 일반적인 시퀀스 모델 백본으로서 맘바는 언어, 오디오, 유전체학과 같은 여러 양식에서 최첨단 성능을 달성합니다. 언어 모델링에서 우리의 맘바-3B 모델은 같은 크기의 트랜스포머를 능가하고 두 배 크기의 트랜스포머와 맞먹는 성능을 보이며, 사전 훈련과 다운스트림 평가 모두에서 성능을 나타납니다.

팁:

  • 맘바는 고전적인 트랜스포머와 견줄 만한 새로운 상태 공간 모델 아키텍처입니다. 이는 구조화된 상태 공간 모델의 발전 선상에 있으며, 플래시어텐션의 정신을 따르는 효율적인 하드웨어 인식 설계와 구현을 특징으로 합니다.
  • 맘바는 어텐션 레이어와 동등한 믹서(mixer) 레이어를 쌓습니다. 맘바의 핵심 로직은 MambaMixer 클래스에 있습니다.
  • 두 가지 구현이 공존합니다: 하나는 최적화되어 빠른 cuda커널을 사용하고, 다른 하나는 단순하지만 모든 장치에서 실행할 수 있습니다!
  • 현재 구현은 원본 cuda커널을 활용합니다: 맘바를 위한 플래시 어텐션의 역할을 하는 것은 mamba-ssm와 causal_conv1d 저장소에 호스팅되어 있습니다. 하드웨어가 지원한다면 반드시 설치하세요!
  • cuda 커널을 최적화하는 방향 보다는, 단순하지만 모든 장치에서 실행가능하도록하는 방향인 ‘단순구현’의 성능을 빠르게 향상시키는 기여를 더 환영하고 있습니다. 🤗

이 모델은 ArthurZ에 의해 기여되었습니다. 원본 코드는 이곳에서 확인할 수 있습니다.

사용

간단한 생성 예제

from transformers import MambaConfig, MambaForCausalLM, AutoTokenizer
import torch

tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-130m-hf")
model = MambaForCausalLM.from_pretrained("state-spaces/mamba-130m-hf")
input_ids = tokenizer("Hey how are you doing?", return_tensors= "pt")["input_ids"]

out = model.generate(input_ids, max_new_tokens=10)
print(tokenizer.batch_decode(out))

Peft 파인튜닝

느린 버전은 학습에서 아주 안정적이진 않습니다. 빠른 버전은 float32가 필요합니다!

from datasets import load_dataset
from trl import SFTTrainer
from peft import LoraConfig
from transformers import AutoTokenizer, AutoModelForCausalLM, TrainingArguments
model_id = "state-spaces/mamba-130m-hf"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)
dataset = load_dataset("Abirate/english_quotes", split="train")
training_args = TrainingArguments(
    output_dir="./results",
    num_train_epochs=3,
    per_device_train_batch_size=4,
    logging_dir='./logs',
    logging_steps=10,
    learning_rate=2e-3
)
lora_config =  LoraConfig(
        r=8,
        target_modules=["x_proj", "embeddings", "in_proj", "out_proj"],
        task_type="CAUSAL_LM",
        bias="none"
)
trainer = SFTTrainer(
    model=model,
    tokenizer=tokenizer,
    args=training_args,
    peft_config=lora_config,
    train_dataset=dataset,
    dataset_text_field="quote",
)
trainer.train()

MambaConfig

class transformers.MambaConfig

< >

( vocab_size = 50280 hidden_size = 768 state_size = 16 num_hidden_layers = 32 layer_norm_epsilon = 1e-05 pad_token_id = 0 bos_token_id = 0 eos_token_id = 0 expand = 2 conv_kernel = 4 use_bias = False use_conv_bias = True hidden_act = 'silu' initializer_range = 0.1 residual_in_fp32 = True time_step_rank = 'auto' time_step_scale = 1.0 time_step_min = 0.001 time_step_max = 0.1 time_step_init_scheme = 'random' time_step_floor = 0.0001 rescale_prenorm_residual = False use_cache = True use_mambapy = False **kwargs )

Parameters

  • vocab_size (int, optional, defaults to 50280) — Vocabulary size of the MAMBA model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling MambaModel.
  • hidden_size (int, optional, defaults to 768) — Dimensionality of the embeddings and hidden states.
  • state_size (int, optional, defaults to 16) — shape of the state space latents.
  • num_hidden_layers (int, optional, defaults to 32) — Number of hidden layers in the model.
  • layer_norm_epsilon (float, optional, defaults to 1e-05) — The epsilon to use in the layer normalization layers.
  • pad_token_id (int, optional, defaults to 0) — Padding token id.
  • bos_token_id (int, optional, defaults to 0) — The id of the beginning of sentence token in the vocabulary.
  • eos_token_id (int, optional, defaults to 0) — The id of the end of sentence token in the vocabulary.
  • expand (int, optional, defaults to 2) — Expanding factor used to determine the intermediate size.
  • conv_kernel (int, optional, defaults to 4) — Size of the convolution kernel.
  • use_bias (bool, optional, defaults to False) — Whether or not to use bias in [“in_proj”, “out_proj”] of the mixer block
  • use_conv_bias (bool, optional, defaults to True) — Whether or not to use bias in the convolution layer of the mixer block.
  • hidden_act (str, optional, defaults to "silu") — The non-linear activation function (function or string) in the decoder.
  • initializer_range (float, optional, defaults to 0.1) — The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
  • residual_in_fp32 (bool, optional, defaults to True) — Whether or not residuals should be in float32. If set to False residuals will keep the same dtype as the rest of the model
  • time_step_rank (Union[int,str], optional, defaults to "auto") — Rank of the discretization projection matrix. "auto" means that it will default to math.ceil(self.hidden_size / 16)
  • time_step_scale (float, optional, defaults to 1.0) — Scale used used to scale dt_proj.bias.
  • time_step_min (float, optional, defaults to 0.001) — Minimum time_step used to bound dt_proj.bias.
  • time_step_max (float, optional, defaults to 0.1) — Maximum time_step used to bound dt_proj.bias.
  • time_step_init_scheme (float, optional, defaults to "random") — Init scheme used for dt_proj.weight. Should be one of ["random","uniform"]
  • time_step_floor (float, optional, defaults to 0.0001) — Minimum clamping value of the dt_proj.bias layer initialization.
  • rescale_prenorm_residual (bool, optional, defaults to False) — Whether or not to rescale out_proj weights when initializing.
  • use_cache (bool, optional, defaults to True) — Whether or not the cache should be used.
  • use_mambapy (bool, optional, defaults to False) — Determines the fallback strategy during training if the CUDA-based official implementation of Mamba is not avaiable. If True, the mamba.py implementation is used. If False, the naive and slower implementation is used. Consider switching to the naive version if memory is limited.

This is the configuration class to store the configuration of a MambaModel. It is used to instantiate a MAMBA model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the MAMBA state-spaces/mamba-2.8b architecture.

Configuration objects inherit from PretrainedConfig and can be used to control the model outputs. Read the documentation from PretrainedConfig for more information.

Example:

>>> from transformers import MambaConfig, MambaModel

>>> # Initializing a Mamba configuration
>>> configuration = MambaConfig()

>>> # Initializing a model (with random weights) from the configuration
>>> model = MambaModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

MambaModel

class transformers.MambaModel

< >

( config )

Parameters

  • config (MambaConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

The bare MAMBA Model transformer outputting raw hidden-states without any specific head on top.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( input_ids: Optional = None inputs_embeds: Optional = None cache_params: Optional = None use_cache: Optional = None output_hidden_states: Optional = None return_dict: Optional = None cache_position: Optional = None attention_mask: Optional = None ) → transformers.models.mamba.modeling_mamba.MambaOutput or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, input_ids_length)) — Indices of input sequence tokens in the vocabulary.

    If cache_params.seqlen_offset>0, only input_ids that do not have their past calculated should be passed as input_ids.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • cache_params (MambaCache, optional) — If passed along, the model uses the previous state in all the blocks (which will give the output for the input_ids provided as if the model add state_input_ids + input_ids as context).
  • use_cache (bool, optional) — If set to True, the cache_params is returned and can be used to quickly generate the next logits.
  • output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
  • return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
  • cache_position (torch.LongTensor of shape (sequence_length), optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.

Returns

transformers.models.mamba.modeling_mamba.MambaOutput or tuple(torch.FloatTensor)

A transformers.models.mamba.modeling_mamba.MambaOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (MambaConfig) and inputs.

  • last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size)) — Sequence of hidden-states at the output of the last layer of the model.

  • cache_params (MambaCache) — The state of the model at the last time step. Can be used in a forward method with the next input_ids to avoid providing the old input_ids.

    Includes both the State space model state matrices after the selective scan, and the Convolutional states

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

The MambaModel forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example:

>>> from transformers import AutoTokenizer, MambaModel
>>> import torch

>>> tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-130m-hf")
>>> model = MambaModel.from_pretrained("state-spaces/mamba-130m-hf")

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> outputs = model(**inputs)

>>> last_hidden_states = outputs.last_hidden_state

MambaLMHeadModel

class transformers.MambaForCausalLM

< >

( config )

Parameters

  • config (MambaConfig) — Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.

The MAMBA Model transformer with a language modeling head on top (linear layer with weights tied to the input embeddings).

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forward

< >

( input_ids: Optional = None attention_mask: Optional = None inputs_embeds: Optional = None cache_params: Optional = None labels: Optional = None output_hidden_states: Optional = None return_dict: Optional = None use_cache: Optional = None cache_position: Optional = None **kwargs ) → transformers.models.mamba.modeling_mamba.MambaCausalLMOutput or tuple(torch.FloatTensor)

Parameters

  • input_ids (torch.LongTensor of shape (batch_size, input_ids_length)) — Indices of input sequence tokens in the vocabulary.

    If cache_params.seqlen_offset>0, only input_ids that do not have their past calculated should be passed as input_ids.

    Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

    What are input IDs?

  • inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) — Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model’s internal embedding lookup matrix.
  • cache_params (MambaCache, optional) — If passed along, the model uses the previous state in all the blocks (which will give the output for the input_ids provided as if the model add state_input_ids + input_ids as context).
  • use_cache (bool, optional) — If set to True, the cache_params is returned and can be used to quickly generate the next logits.
  • output_hidden_states (bool, optional) — Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.
  • return_dict (bool, optional) — Whether or not to return a ModelOutput instead of a plain tuple.
  • cache_position (torch.LongTensor of shape (sequence_length), optional) — Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.
  • labels (torch.LongTensor of shape (batch_size, sequence_length), optional) — Labels for language modeling. Note that the labels are shifted inside the model, i.e. you can set labels = input_ids Indices are selected in [-100, 0, ..., config.vocab_size] All labels set to -100 are ignored (masked), the loss is only computed for labels in [0, ..., config.vocab_size]

Returns

transformers.models.mamba.modeling_mamba.MambaCausalLMOutput or tuple(torch.FloatTensor)

A transformers.models.mamba.modeling_mamba.MambaCausalLMOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (MambaConfig) and inputs.

  • loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) — Language modeling loss (for next-token prediction).

  • logits (torch.FloatTensor of shape (batch_size, sequence_length, config.vocab_size)) — Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

  • cache_params (MambaCache) — The state of the model at the last time step. Can be used in a forward method with the next input_ids to avoid providing the old input_ids.

    Includes both the State space model state matrices after the selective scan, and the Convolutional states

  • hidden_states (tuple(torch.FloatTensor), optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) — Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

    Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

The MambaForCausalLM forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example:

>>> import torch
>>> from transformers import AutoTokenizer, MambaForCausalLM

>>> tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-130m-hf")
>>> model = MambaForCausalLM.from_pretrained("state-spaces/mamba-130m-hf")

>>> inputs = tokenizer("Hello, my dog is cute", return_tensors="pt")
>>> outputs = model(**inputs, labels=inputs["input_ids"])
>>> loss = outputs.loss
>>> logits = outputs.logits
< > Update on GitHub

←(번역중) M2M100 Mamba2→