from typing import TYPE_CHECKING, Any, Mapping, Optional, OrderedDict from packaging import version from transformers.auto.configuration_auto import AutoConfig from transformers.configuration_utils import PretrainedConfig from transformers.utils import logging if TYPE_CHECKING: from ... import PreTrainedTokenizerBase, TensorType logger = logging.get_logger(__name__) """ Mistral model configuration""" MISTRAL_PRETRAINED_CONFIG_ARCHIVE_MAP = { "mistralai/Mistral-7B-v0.1": "https://huggingface.co/mistralai/Mistral-7B-v0.1/resolve/main/config.json", "mistralai/Mistral-7B-Instruct-v0.1": "https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1/resolve/main/config.json", } class EncoderDecoderConfig(PretrainedConfig): is_composition = True def __init__(self, **kwargs): super().__init__(**kwargs) if "encoder" not in kwargs or "decoder" not in kwargs: raise ValueError( f"A configuraton of type {self.model_type} cannot be instantiated because " f"not both `encoder` and `decoder` sub-configurations are passed, but only {kwargs}" ) encoder_config = kwargs.pop("encoder") encoder_model_type = encoder_config.pop("model_type") decoder_config = kwargs.pop("decoder") decoder_model_type = decoder_config.pop("model_type") self.encoder = AutoConfig.for_model(encoder_model_type, **encoder_config) self.decoder = AutoConfig.for_model(decoder_model_type, **decoder_config) self.is_encoder_decoder = True @classmethod def from_encoder_decoder_configs( cls, encoder_config: PretrainedConfig, decoder_config: PretrainedConfig, **kwargs ) -> PretrainedConfig: r""" Instantiate a [`SpeechEncoderDecoderConfig`] (or a derived class) from a pre-trained encoder model configuration and decoder model configuration. Returns: [`SpeechEncoderDecoderConfig`]: An instance of a configuration object """ logger.info("Setting `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config") decoder_config.is_decoder = True decoder_config.add_cross_attention = True return cls(encoder=encoder_config.to_dict(), decoder=decoder_config.to_dict(), **kwargs) class VisionEncoderDecoderConfig(PretrainedConfig): r""" [`VisionEncoderDecoderConfig`] is the configuration class to store the configuration of a [`VisionEncoderDecoderModel`]. It is used to instantiate a Vision-Encoder-Text-Decoder model according to the specified arguments, defining the encoder and decoder configs. Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: kwargs (*optional*): Dictionary of keyword arguments. Notably: - **encoder** ([`PretrainedConfig`], *optional*) -- An instance of a configuration object that defines the encoder config. - **decoder** ([`PretrainedConfig`], *optional*) -- An instance of a configuration object that defines the decoder config. Examples: ```python >>> from transformers import BertConfig, ViTConfig, VisionEncoderDecoderConfig, VisionEncoderDecoderModel >>> # Initializing a ViT & BERT style configuration >>> config_encoder = ViTConfig() >>> config_decoder = BertConfig() >>> config = VisionEncoderDecoderConfig.from_encoder_decoder_configs(config_encoder, config_decoder) >>> # Initializing a ViTBert model (with random weights) from a ViT & google-bert/bert-base-uncased style configurations >>> model = VisionEncoderDecoderModel(config=config) >>> # Accessing the model configuration >>> config_encoder = model.config.encoder >>> config_decoder = model.config.decoder >>> # set decoder config to causal lm >>> config_decoder.is_decoder = True >>> config_decoder.add_cross_attention = True >>> # Saving the model, including its configuration >>> model.save_pretrained("my-model") >>> # loading model and config from pretrained folder >>> encoder_decoder_config = VisionEncoderDecoderConfig.from_pretrained("my-model") >>> model = VisionEncoderDecoderModel.from_pretrained("my-model", config=encoder_decoder_config) ```""" model_type = "vision-encoder-decoder" is_composition = True def __init__(self, **kwargs): super().__init__(**kwargs) if "encoder" not in kwargs or "decoder" not in kwargs: raise ValueError( f"A configuraton of type {self.model_type} cannot be instantiated because " f"not both `encoder` and `decoder` sub-configurations are passed, but only {kwargs}" ) encoder_config = kwargs.pop("encoder") encoder_model_type = encoder_config.pop("model_type") decoder_config = kwargs.pop("decoder") decoder_model_type = decoder_config.pop("model_type") self.encoder = AutoConfig.for_model(encoder_model_type, **encoder_config) self.decoder = AutoConfig.for_model(decoder_model_type, **decoder_config) self.is_encoder_decoder = True @classmethod def from_encoder_decoder_configs( cls, encoder_config: PretrainedConfig, decoder_config: PretrainedConfig, **kwargs ) -> PretrainedConfig: r""" Instantiate a [`VisionEncoderDecoderConfig`] (or a derived class) from a pre-trained encoder model configuration and decoder model configuration. Returns: [`VisionEncoderDecoderConfig`]: An instance of a configuration object """ logger.info("Setting `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config") decoder_config.is_decoder = True decoder_config.add_cross_attention = True return cls(encoder=encoder_config.to_dict(), decoder=decoder_config.to_dict(), **kwargs) class SpeechEncoderDecoderConfig(PretrainedConfig): r""" [`SpeechEncoderDecoderConfig`] is the configuration class to store the configuration of a [`SpeechEncoderDecoderModel`]. It is used to instantiate an Encoder Decoder model according to the specified arguments, defining the encoder and decoder configs. Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: kwargs (*optional*): Dictionary of keyword arguments. Notably: - **encoder** ([`PretrainedConfig`], *optional*) -- An instance of a configuration object that defines the encoder config. - **decoder** ([`PretrainedConfig`], *optional*) -- An instance of a configuration object that defines the decoder config. Examples: ```python >>> from transformers import BertConfig, Wav2Vec2Config, SpeechEncoderDecoderConfig, SpeechEncoderDecoderModel >>> # Initializing a Wav2Vec2 & BERT style configuration >>> config_encoder = Wav2Vec2Config() >>> config_decoder = BertConfig() >>> config = SpeechEncoderDecoderConfig.from_encoder_decoder_configs(config_encoder, config_decoder) >>> # Initializing a Wav2Vec2Bert model from a Wav2Vec2 & google-bert/bert-base-uncased style configurations >>> model = SpeechEncoderDecoderModel(config=config) >>> # Accessing the model configuration >>> config_encoder = model.config.encoder >>> config_decoder = model.config.decoder >>> # set decoder config to causal lm >>> config_decoder.is_decoder = True >>> config_decoder.add_cross_attention = True >>> # Saving the model, including its configuration >>> model.save_pretrained("my-model") >>> # loading model and config from pretrained folder >>> encoder_decoder_config = SpeechEncoderDecoderConfig.from_pretrained("my-model") >>> model = SpeechEncoderDecoderModel.from_pretrained("my-model", config=encoder_decoder_config) ```""" model_type = "speech-encoder-decoder" is_composition = True def __init__(self, **kwargs): super().__init__(**kwargs) if "encoder" not in kwargs or "decoder" not in kwargs: raise ValueError( f"A configuraton of type {self.model_type} cannot be instantiated because not both `encoder` and" f" `decoder` sub-configurations are passed, but only {kwargs}" ) encoder_config = kwargs.pop("encoder") encoder_model_type = encoder_config.pop("model_type") decoder_config = kwargs.pop("decoder") decoder_model_type = decoder_config.pop("model_type") self.encoder = AutoConfig.for_model(encoder_model_type, **encoder_config) self.decoder = AutoConfig.for_model(decoder_model_type, **decoder_config) self.is_encoder_decoder = True @classmethod def from_encoder_decoder_configs( cls, encoder_config: PretrainedConfig, decoder_config: PretrainedConfig, **kwargs ) -> PretrainedConfig: r""" Instantiate a [`SpeechEncoderDecoderConfig`] (or a derived class) from a pre-trained encoder model configuration and decoder model configuration. Returns: [`SpeechEncoderDecoderConfig`]: An instance of a configuration object """ logger.info("Setting `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config") decoder_config.is_decoder = True decoder_config.add_cross_attention = True return cls(encoder=encoder_config.to_dict(), decoder=decoder_config.to_dict(), **kwargs) class MistralConfig(PretrainedConfig): is_composition = True r""" This is the configuration class to store the configuration of a [`MistralModel`]. It is used to instantiate an Mistral 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 Mistral-7B-v0.1 or Mistral-7B-Instruct-v0.1. [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) [mistralai/Mistral-7B-Instruct-v0.1](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: vocab_size (`int`, *optional*, defaults to 32000): Vocabulary size of the Mistral model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`MistralModel`] hidden_size (`int`, *optional*, defaults to 4096): Dimension of the hidden representations. intermediate_size (`int`, *optional*, defaults to 14336): Dimension of the MLP representations. num_hidden_layers (`int`, *optional*, defaults to 32): Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 32): Number of attention heads for each attention layer in the Transformer encoder. num_key_value_heads (`int`, *optional*, defaults to 8): This is the number of key_value heads that should be used to implement Grouped Query Attention. If `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if `num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details checkout [this paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to `8`. hidden_act (`str` or `function`, *optional*, defaults to `"silu"`): The non-linear activation function (function or string) in the decoder. max_position_embeddings (`int`, *optional*, defaults to `4096*32`): The maximum sequence length that this model might ever be used with. Mistral's sliding window attention allows sequence of up to 4096*32 tokens. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. rms_norm_eps (`float`, *optional*, defaults to 1e-06): The epsilon used by the rms normalization layers. use_cache (`bool`, *optional*, defaults to `True`): Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if `config.is_decoder=True`. pad_token_id (`int`, *optional*): The id of the padding token. bos_token_id (`int`, *optional*, defaults to 1): The id of the "beginning-of-sequence" token. eos_token_id (`int`, *optional*, defaults to 2): The id of the "end-of-sequence" token. tie_word_embeddings (`bool`, *optional*, defaults to `False`): Whether the model's input and output word embeddings should be tied. rope_theta (`float`, *optional*, defaults to 10000.0): The base period of the RoPE embeddings. sliding_window (`int`, *optional*, defaults to 4096): Sliding window attention window size. If not specified, will default to `4096`. attention_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for the attention probabilities. ```python >>> from transformers import MistralModel, MistralConfig >>> # Initializing a Mistral 7B style configuration >>> configuration = MistralConfig() >>> # Initializing a model from the Mistral 7B style configuration >>> model = MistralModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```""" model_type = ["mistral","speech-encoder-decoder","image-encoder-decoder","mistral-encoder-decoder"] # model_type = "mistral-encoder-decoder" keys_to_ignore_at_inference = ["past_key_values"] def __init__( self, vocab_size=32000, hidden_size=4096, intermediate_size=14336, num_hidden_layers=32, num_attention_heads=32, num_key_value_heads=8, hidden_act="silu", max_position_embeddings=4096 * 32, initializer_range=0.02, rms_norm_eps=1e-6, use_cache=True, pad_token_id=None, bos_token_id=1, eos_token_id=2, tie_word_embeddings=False, sliding_window=4096, attention_dropout=0.0, # for yarn Later rope_theta=10000.0, rope_scaling=None, # for thought generation Later max_thoughts=16, max_temperature=10, complexity_factor = 0.5, merged_talk_heads=True, merged_lm_and_talk_heads=False, merged_lm_and_think_heads=True, use_concat_talk_head=True, use_shallow_think=True, use_shallow_talk=False, use_complex_think_head=False, use_complex_talk_head=True, use_weighted_talk_head=True, hidden_dropout_prob=0.00, **kwargs, ): super().__init__( pad_token_id=pad_token_id, bos_token_id=bos_token_id, eos_token_id=eos_token_id, tie_word_embeddings=tie_word_embeddings, **kwargs, ) self.vocab_size = vocab_size self.max_position_embeddings = max_position_embeddings self.hidden_size = hidden_size self.intermediate_size = intermediate_size self.num_hidden_layers = num_hidden_layers self.num_attention_heads = num_attention_heads self.sliding_window = sliding_window # for backward compatibility if num_key_value_heads is None: num_key_value_heads = num_attention_heads self.num_key_value_heads = num_key_value_heads self.hidden_act = hidden_act self.initializer_range = initializer_range self.rms_norm_eps = rms_norm_eps self.use_cache = use_cache self.attention_dropout = attention_dropout # yarn self.rope_scaling = rope_scaling self._rope_scaling_validation() self.rope_theta = rope_theta #Thought gen self.max_thoughts = max_thoughts self.complexity_factor = complexity_factor self.max_temperature = max_temperature self.merged_talk_heads = merged_talk_heads self.merged_lm_and_talk_heads = merged_lm_and_talk_heads self.merged_lm_and_think_heads = merged_lm_and_think_heads self.use_concat_talk_head = use_concat_talk_head self.use_shallow_think = use_shallow_think self.use_shallow_talk = use_shallow_talk self.use_complex_think_head = use_complex_think_head self.use_complex_talk_head = use_complex_talk_head self.use_weighted_talk_head = use_weighted_talk_head self.hidden_dropout_prob = hidden_dropout_prob #Encoder Decoder - Currently only a single EncoderDecoder is supported -Later will be eXpanded to suport both models if "encoder" not in kwargs or "decoder" not in kwargs: raise ValueError( f"A configuraton of type {self.model_type} cannot be instantiated because " f"not both `encoder` and `decoder` sub-configurations are passed, but only {kwargs}" ) encoder_config = kwargs.pop("encoder") encoder_model_type = encoder_config.pop("model_type") decoder_config = kwargs.pop("decoder") decoder_model_type = decoder_config.pop("model_type") self.encoder = AutoConfig.for_model(encoder_model_type, **encoder_config) self.decoder = AutoConfig.for_model(decoder_model_type, **decoder_config) self.is_encoder_decoder = True @classmethod def from_encoder_decoder_configs( cls, encoder_config: PretrainedConfig, decoder_config: PretrainedConfig, **kwargs ) -> PretrainedConfig: r""" Instantiate a [`SpeechEncoderDecoderConfig`] (or a derived class) from a pre-trained encoder model configuration and decoder model configuration. Returns: [`SpeechEncoderDecoderConfig`]: An instance of a configuration object """ logger.info("Setting `config.is_decoder=True` and `config.add_cross_attention=True` for decoder_config") decoder_config.is_decoder = True decoder_config.add_cross_attention = True return cls(encoder=encoder_config.to_dict(), decoder=decoder_config.to_dict(), **kwargs) def _rope_scaling_validation(self): """ Validate the `rope_scaling` configuration. """ if self.rope_scaling is None: return if not isinstance(self.rope_scaling, dict): raise ValueError( "`rope_scaling` must be a dictionary, " f"got {self.rope_scaling}" ) rope_scaling_type = self.rope_scaling.get("type", None) rope_scaling_factor = self.rope_scaling.get("factor", None) if rope_scaling_type is None or rope_scaling_type not in ["linear", "dynamic", "yarn", "dynamic-yarn"]: raise ValueError( f"`rope_scaling`'s name field must be one of ['linear', 'dynamic', 'yarn', 'dynamic-yarn'], got {rope_scaling_type}" ) if rope_scaling_factor is None or not isinstance(rope_scaling_factor, float) or rope_scaling_factor <= 1.0: raise ValueError(f"`rope_scaling`'s factor field must be an float > 1, got {rope_scaling_factor}") if rope_scaling_type == "yarn" or rope_scaling_type == "dynamic-yarn": original_max_position_embeddings = self.rope_scaling.get("original_max_position_embeddings", None) if original_max_position_embeddings is None or not isinstance(original_max_position_embeddings, int): raise ValueError(f"`rope_scaling.original_max_position_embeddings` must be set to an int when using yarn, and dynamic-yarn")