`iV. ddlmZddlZddlmZddlmZmZddlmZddl m Z m Z ddl m Z dd lmZdd lmZmZdd lmZd d lmZd dlmZmZmZmZmZmZmZmZm Z m!Z!m"Z"GddeeZ#GddeZ$GddeZ%Gdde!Z&GddeZ'Gdde Z(Gdde"Z)GddeZ*Gdd eZ+Gd!d"eZ,Gd#d$eZ-Gd%d&eZ.gd'Z/dS)()OptionalN)nn)Cache DynamicCache)PretrainedConfig)create_causal_mask!create_sliding_window_causal_mask)BaseModelOutputWithPast)Unpack)TransformersKwargsauto_docstring)check_model_inputs) MistralConfig) Qwen2AttentionQwen2DecoderLayerQwen2ForCausalLMQwen2ForQuestionAnsweringQwen2ForSequenceClassificationQwen2ForTokenClassificationQwen2MLP Qwen2ModelQwen2PreTrainedModel Qwen2RMSNormQwen2RotaryEmbeddingcFeZdZdZdZ ddZdS)MinistralConfiga This is the configuration class to store the configuration of a [`MinistralModel`]. It is used to instantiate an Ministral 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 Ministral-8B-Instruct-2410. [mistralai/Ministral-8B-Instruct-2410](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410) [mistralai/Ministral-8B-Instruct-2410](https://huggingface.co/mistralai/Ministral-8B-Instruct-2410) 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 Ministral model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`MinistralModel`] 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, check out [this paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `8`. head_dim (`int`, *optional*, defaults to `hidden_size // num_attention_heads`): The attention head dimension. 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. Ministral'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. layer_types (`list`, *optional*): Attention pattern for each layer. ```python >>> from transformers import MinistralModel, MinistralConfig >>> # Initializing a Ministral 8B style configuration >>> configuration = MinistralConfig() >>> # Initializing a model from the Ministral 8B style configuration >>> model = MinistralModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` ministral}8 Nsilu{Gz?ư>TrF@c Vtj|f| |||d|||_| |_||_||_||_||_||_||_ ||}||_ ||_ | |_ | |_ | |_||_||_||_|j|jdndg|z|_dSdS)N) pad_token_id bos_token_id eos_token_idtie_word_embeddingssliding_attentionfull_attention)r__init__ vocab_sizemax_position_embeddings hidden_sizeintermediate_sizenum_hidden_layersnum_attention_headssliding_windowhead_dimnum_key_value_heads hidden_actinitializer_range rms_norm_eps use_cache rope_thetaattention_dropout layer_types)selfr4r6r7r8r9r<r;r=r5r>r?r@r-r.r/r0rAr:rBrCkwargss /home/jaya/work/projects/VOICE-AGENT/VIET/agent-env/lib/python3.11/site-packages/transformers/models/ministral/modular_ministral.pyr3zMinistralConfig.__init__ks0 !  %%% 3       %'>$&!2!2#6 ,    &"5 #6 $!2("$!2&   #'+':'F##L\ ! "D    $ #)r r!r"r#r#r$Nr%r&r'r(TNr)rFr*r!r+N)__name__ __module__ __qualname____doc__ model_typer3rGrFrrszI I VJ )!+9"9"9"9"9"9"rGrceZdZdS) MinistralMLPNrHrIrJrMrGrFrOrODrGrOc$eZdZdeffd ZxZS)MinistralAttention layer_idxc`t||tj|j|j|jzd|_tj|j|j|jzd|_ tj|j|j|jzd|_ dS)NF)bias) superr3rLinearr6r9r;q_projr<k_projv_proj)rDconfigrT __class__s rFr3zMinistralAttention.__init__s +++i 2F4NQUQ^4^ejkkk i 2F4NQUQ^4^ejkkk i 2F4NQUQ^4^ejkkk rG)rHrIrJintr3 __classcell__r]s@rFrSrSsOl#llllllllllrGrSceZdZdS)MinistralRMSNormNrPrMrGrFrbrbrQrGrbceZdZdS)MinistralDecoderLayerNrPrMrGrFrdrdrQrGrdceZdZdS)MinistralPreTrainedModelNrPrMrGrFrfrfrQrGrfceZdZdS)MinistralRotaryEmbeddingNrPrMrGrFrhrhrQrGrhceZdZdeffd Zee ddeej deej deej dee deej d ee d eej d eed efd ZxZS)MinistralModelr\cNt||`dS)N)rWr3has_sliding_layers)rDr\r]s rFr3zMinistralModel.__init__s'     # # #rGN input_idsattention_mask position_idspast_key_values inputs_embedsr@cache_positionrEreturnc |du|duz rtd|||}|r|t|j}|B||nd} t j| | |jdz|j}|| d}t|x} ts'|j|||||d} td i| td i| d} |} || |} |jd|jjD]}|| f| |j||||| d|} || } t'| |r|nd S) Nz:You must specify exactly one of input_ids or inputs_embeds)r\rr))device)r\ input_embedsrnrrrpro)r2r1)rnrorpr@rrposition_embeddings)last_hidden_staterprM) ValueError embed_tokensrr\get_seq_lengthtorcharangeshaperu unsqueeze isinstancedictr r rotary_emblayersr8attention_typenormr )rDrmrnrorprqr@rrrEpast_seen_tokenscausal_mask_mapping mask_kwargs hidden_statesrw decoder_layers rFforwardzMinistralModel.forwards  -t"; < [YZZ Z   --i88M  ?0*$+>>>O  !CRC^==???de "\ "2]5H5K"KTaThN  )33A66L?-FF + -"0"0#2 , K#5"C"C{"C"C%F%U%U%U%U##  & #oom\JJ![)H4;+H)HI  M)M 2=3OP) /#-$7    MM -00 &+/8BOOd    rG)NNNNNNN)rHrIrJrr3rrrr| LongTensorTensorr FloatTensorboolr r r rr_r`s@rFrjrjs*$$$$$$$151537+/59$(59C C E,-C !.C u/0 C "% C   12 C D>C !!12C +,C  !C C C ^C C C C C rGrjceZdZdS)MinistralForCausalLMNrPrMrGrFrrrQrGrceZdZdS)"MinistralForSequenceClassificationNrPrMrGrFrrrQrGrceZdZdS)MinistralForTokenClassificationNrPrMrGrFrrrQrGrceZdZdS)MinistralForQuestionAnsweringNrPrMrGrFrrrQrGr)rrfrjrrrr)0typingrr|r cache_utilsrrconfiguration_utilsr masking_utilsr r modeling_outputsr processing_utilsr utilsr r utils.genericrmistral.configuration_mistralrqwen2.modeling_qwen2rrrrrrrrrrrrrOrSrbrdrfrhrjrrrr__all__rMrGrFrs ........333333RRRRRRRR777777&&&&&&77777777//////999999                          G"G"G"G"G"m%5G"G"G"T     8   llllllll     |        -        3        3   J J J J J ZJ J J Z     +        )G        &A        $=      rG