
    rhS                     h   d dl mZmZmZ d dlZd dlmZ ddlmZ ddlm	Z	m
Z
 ddlmZ ddlmZ dd	lmZmZ dd
lmZ ddlmZmZmZmZ ddlmZmZ ddlmZmZ ddlm Z m!Z! ddl"m#Z# ddl$m%Z%m&Z&m'Z' ddl(m)Z) ddl*m+Z+  G d dejX                        Z-d Z.d;dZ/dej`                  de1dej`                  fdZ2	 d<dejX                  dej`                  dej`                  dej`                  d eej`                     d!e3d"e3d#e#e%   fd$Z4 G d% d&ejX                        Z5 ed'       G d( d)ejX                               Z6 G d* d+e      Z7e& G d, d-e!             Z8 G d. d/ejX                        Z9e& G d0 d1e8             Z:e& G d2 d3e8e             Z; G d4 d5ee8      Z< G d6 d7ee8      Z= G d8 d9ee8      Z>g d:Z?y)=    )CallableOptionalUnionN)nn   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hub)create_causal_mask!create_sliding_window_causal_mask)FlashAttentionKwargs)GenericForQuestionAnswering GenericForSequenceClassificationGenericForTokenClassificationGradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)TransformersKwargsauto_docstringcan_return_tuple)check_model_inputs   )Qwen2Configc                   $     e Zd Z fdZd Z xZS )Qwen2MLPc                    t         |           || _        |j                  | _        |j                  | _        t        j                  | j                  | j                  d      | _        t        j                  | j                  | j                  d      | _        t        j                  | j                  | j                  d      | _	        t        |j                     | _        y NFbias)super__init__confighidden_sizeintermediate_sizer   Linear	gate_projup_proj	down_projr   
hidden_actact_fnselfr)   	__class__s     {/var/www/html/ai-insurance-compliance-backend/venv/lib/python3.12/site-packages/transformers/models/qwen2/modeling_qwen2.pyr(   zQwen2MLP.__init__"   s    !--!'!9!94#3#3T5K5KRWXyy!1!143I3IPUV4#9#94;K;KRWXV../    c                     | j                  | j                  | j                  |            | j                  |      z        }|S N)r/   r1   r-   r.   )r3   xr/   s      r5   forwardzQwen2MLP.forward,   s6    NN4;;t~~a/@#ADLLQRO#ST	r6   )__name__
__module____qualname__r(   r:   __classcell__r4   s   @r5   r"   r"   !   s    0r6   r"   c                     | dd| j                   d   dz  f   }| d| j                   d   dz  df   }t        j                  | |fd      S )z*Rotates half the hidden dims of the input..N   dim)shapetorchcat)r9   x1x2s      r5   rotate_halfrJ   1   sZ    	
3"!''"+"""	#B	
3q ""	#B99rc2YB''r6   c                     |j                  |      }|j                  |      }| |z  t        |       |z  z   }||z  t        |      |z  z   }||fS )a  Applies Rotary Position Embedding to the query and key tensors.

    Args:
        q (`torch.Tensor`): The query tensor.
        k (`torch.Tensor`): The key tensor.
        cos (`torch.Tensor`): The cosine part of the rotary embedding.
        sin (`torch.Tensor`): The sine part of the rotary embedding.
        position_ids (`torch.Tensor`, *optional*):
            Deprecated and unused.
        unsqueeze_dim (`int`, *optional*, defaults to 1):
            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
    Returns:
        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
    )	unsqueezerJ   )qkcossinposition_idsunsqueeze_dimq_embedk_embeds           r5   apply_rotary_pos_embrU   8   sY    ( --
&C
--
&C3w;q>C/0G3w;q>C/0GGr6   hidden_statesn_repreturnc                     | j                   \  }}}}|dk(  r| S | dddddddddf   j                  |||||      } | j                  |||z  ||      S )z
    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
    r   N)rE   expandreshape)rV   rW   batchnum_key_value_headsslenhead_dims         r5   	repeat_kvr`   S   so    
 2?1D1D.Ehz!!Qa"23::5BUW\^bdlmM  (;e(CT8TTr6   modulequerykeyvalueattention_maskscalingdropoutkwargsc                 T   t        || j                        }t        || j                        }	t        j                  ||j	                  dd            |z  }
|#|d d d d d d d |j
                  d   f   }|
|z   }
t        j                  j                  |
dt        j                        j                  |j                        }
t        j                  j                  |
|| j                        }
t        j                  |
|	      }|j	                  dd      j                         }||
fS )NrB   r   rA   )rD   dtype)ptrainingr   )r`   num_key_value_groupsrF   matmul	transposerE   r   
functionalsoftmaxfloat32tork   rg   rm   
contiguous)ra   rb   rc   rd   re   rf   rg   rh   
key_statesvalue_statesattn_weightscausal_maskattn_outputs                r5   eager_attention_forwardr{   _   s    3 ; ;<JUF$?$?@L<<z';';Aq'ABWLL!$Q1.D
0@0@0D.D%DE#k1==((2U]](SVVW\WbWbcL==((6??([L,,|\:K''1-88:K$$r6   c                   6    e Zd ZdZdedef fdZ	 	 ddej                  de	ej                  ej                  f   de
ej                     de
e   d	e
ej                     d
ee   de	ej                  e
ej                     e
e	ej                        f   fdZ xZS )Qwen2Attentionz=Multi-headed attention from 'Attention Is All You Need' paperr)   	layer_idxc                 j   t         |           || _        || _        t	        |d|j
                  |j                  z        | _        |j                  |j                  z  | _	        | j                  dz  | _
        |j                  | _        d| _        t        j                  |j
                  |j                  | j                  z  d      | _        t        j                  |j
                  |j                  | j                  z  d      | _        t        j                  |j
                  |j                  | j                  z  d      | _        t        j                  |j                  | j                  z  |j
                  d      | _        |j&                  |   dk(  r|j(                  | _        y d | _        y )Nr_   g      Tr%   Fsliding_attention)r'   r(   r)   r~   getattrr*   num_attention_headsr_   r]   rn   rf   attention_dropout	is_causalr   r,   q_projk_projv_projo_projlayer_typessliding_windowr3   r)   r~   r4   s      r5   r(   zQwen2Attention.__init__|   sU   "
F4F4F&JdJd4de$*$>$>&B\B\$\!}}d*!'!9!9ii 2 2F4N4NQUQ^Q^4^eijii 2 2F4N4NQUQ^Q^4^eijii 2 2F4N4NQUQ^Q^4^eijii : :T]] JFL^L^ejk7=7I7I)7TXk7kf33qur6   rV   position_embeddingsre   past_key_valuecache_positionrh   rX   c                 J   |j                   d d }g |d| j                  }| j                  |      j                  |      j	                  dd      }	| j                  |      j                  |      j	                  dd      }
| j                  |      j                  |      j	                  dd      }|\  }}t        |	|
||      \  }	}
|'|||d}|j                  |
|| j                  |      \  }
}t        }| j                  j                  dk7  rt        | j                  j                     } || |	|
||f| j                  sdn| j                  | j                   | j"                  d|\  }} |j$                  g |d j'                         }| j)                  |      }||fS )NrA   r   rB   )rP   rO   r   eager        )rg   rf   r   )rE   r_   r   viewrp   r   r   rU   updater~   r{   r)   _attn_implementationr   rm   r   rf   r   r[   ru   r   )r3   rV   r   re   r   r   rh   input_shapehidden_shapequery_statesrv   rw   rO   rP   cache_kwargsattention_interfacerz   rx   s                     r5   r:   zQwen2Attention.forward   s    $))#2.88b8$--8{{=166|DNNqRST[[/44\BLLQPQR
{{=166|DNNqRST&S#7jRUWZ#[ j%#&snUL'5'<'<ZW[WeWegs't$J(?;;++w6"9$++:Z:Z"[$7
%
  $}}C$2H2HLL..
%
 
%
!\ *k));;;;FFHkk+.L((r6   )NN)r;   r<   r=   __doc__r    intr(   rF   Tensortupler   r	   
LongTensorr   r   r:   r>   r?   s   @r5   r}   r}   y   s    Gv{ vs v( +/59*)||*) #5<<#=>*) !.	*)
 !*) !!1!12*) -.*) 
u||Xell3XeELL>Q5RR	S*)r6   r}   RMSNormc                   ,     e Zd Zd fd	Zd Zd Z xZS )Qwen2RMSNormc                     t         |           t        j                  t	        j
                  |            | _        || _        y)z;
        Qwen2RMSNorm is equivalent to T5LayerNorm
        N)r'   r(   r   	ParameterrF   onesweightvariance_epsilon)r3   r*   epsr4   s      r5   r(   zQwen2RMSNorm.__init__   s1     	ll5::k#:; #r6   c                 "   |j                   }|j                  t        j                        }|j	                  d      j                  dd      }|t        j                  || j                  z         z  }| j                  |j                  |      z  S )NrB   rA   T)keepdim)	rk   rt   rF   rs   powmeanrsqrtr   r   )r3   rV   input_dtypevariances       r5   r:   zQwen2RMSNorm.forward   sy    #))%((7 $$Q',,R,>%Ht?T?T4T(UU{{]--k:::r6   c                 ^    t        | j                  j                         d| j                   S )Nz, eps=)r   r   rE   r   r3   s    r5   
extra_reprzQwen2RMSNorm.extra_repr   s*    ))*+6$2G2G1HIIr6   )gư>)r;   r<   r=   r(   r:   r   r>   r?   s   @r5   r   r      s    $;Jr6   r   c                   (    e Zd Zdedef fdZ	 	 	 	 	 	 ddej                  deej                     deej                     dee
   dee   d	eej                     d
eeej                  ej                  f      dee   deej                     fdZ xZS )Qwen2DecoderLayerr)   r~   c                 H   t         |           |j                  | _        t        ||      | _        t        |      | _        t        |j                  |j                        | _	        t        |j                  |j                        | _
        |j                  |   | _        y )N)r)   r~   r   )r'   r(   r*   r}   	self_attnr"   mlpr   rms_norm_epsinput_layernormpost_attention_layernormr   attention_typer   s      r5   r(   zQwen2DecoderLayer.__init__   s    !--'vKF#+F,>,>FDWDWX(4V5G5GVM`M`(a%$00;r6   rV   re   rQ   r   	use_cacher   r   rh   rX   c                     |}	| j                  |      } | j                  d|||||||d|\  }}
|	|z   }|}	| j                  |      }| j                  |      }|	|z   }|S )N)rV   re   rQ   r   r   r   r    )r   r   r   r   )r3   rV   re   rQ   r   r   r   r   rh   residual_s              r5   r:   zQwen2DecoderLayer.forward   s     !,,];)4>> 	
')%)) 3	
 	
q !=0 !55mD/ =0r6   )NNNFNN)r;   r<   r=   r    r   r(   rF   r   r   r   r	   boolr   r   r   r:   r>   r?   s   @r5   r   r      s    	<{ 	<s 	< 2637*.$)59KO|| !. u//0	
 ! D> !!1!12 &eELL%,,,F&GH +, 
u||	r6   r   c                   J    e Zd ZU eed<   dZdZdgZdgZdZ	dZ
dZdZdZeedZy)Qwen2PreTrainedModelr)   modelTr   past_key_values)rV   
attentionsN)r;   r<   r=   r    __annotations__base_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_supports_sdpa_supports_flex_attn_can_compile_fullgraph_supports_attention_backendr   r}   _can_record_outputsr   r6   r5   r   r      sQ    &*#,-#4"5N!"&*$r6   r   c                   ^     e Zd Zddef fdZ ej                         ed               Z xZ	S )Qwen2RotaryEmbeddingr)   c                    t         |           t        |d      rUt        |j                  t
              r;|j                  j                  d|j                  j                  d            | _        nd| _        |j                  | _	        |j                  | _
        || _        t        | j                     | _        | j                  | j                  |      \  }| _        | j                  d|d       | j                   | _        y )Nrope_scaling	rope_typetypedefaultinv_freqF)
persistent)r'   r(   hasattr
isinstancer   dictgetr   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenr)   r   rope_init_fnattention_scalingregister_bufferr   original_inv_freq)r3   r)   devicer   r4   s       r5   r(   zQwen2RotaryEmbedding.__init__  s    6>*z&:M:Mt/T#0044[&BUBUBYBYZ`BabDN&DN"("@"@$*$B$B!/?+/+<+<T[[&+Q($(ZeD!%r6   c                 b   | j                   d d d d f   j                         j                  |j                  d   dd      j	                  |j
                        }|d d d d d f   j                         }t        |j
                  j                  t              r/|j
                  j                  dk7  r|j
                  j                  nd}t        j                  |d      5  |j                         |j                         z  j                  dd      }t        j                  ||fd	      }|j                         | j                  z  }|j                         | j                  z  }	d d d        j	                  |j                   
      	j	                  |j                   
      fS # 1 sw Y   AxY w)Nr   rA   r   mpscpuF)device_typeenabledrB   rC   )rk   )r   floatrZ   rE   rt   r   r   r   strrF   autocastrp   rG   rO   r   rP   rk   )
r3   r9   rQ   inv_freq_expandedposition_ids_expandedr   freqsembrO   rP   s
             r5   r:   zQwen2RotaryEmbedding.forward   sV    !MM$4-8>>@GGHZHZ[\H]_acdehhijiqiqr ,QaZ 8 > > @'1!((--'E!((--[`J`ahhmmfk^^UC 	5&,,.1F1L1L1NNYYZ[]^_E))UEN3C'')d444C'')d444C		5 vvAGGv$cff177f&;;;	5 	5s    BF%%F.r8   )
r;   r<   r=   r    r(   rF   no_gradr   r:   r>   r?   s   @r5   r   r     s3    /{ /" U]]_<  <r6   r   c                       e Zd Zdef fdZee	 	 	 	 	 	 	 ddeej                     deej                     deej                     dee   deej                     dee   d	eej                     d
ee   defd              Z xZS )
Qwen2Modelr)   c           	      F   t         |   |       |j                  | _        |j                  | _        t        j                  |j                  |j                  | j                        | _        t        j                  t        |j                        D cg c]  }t        ||       c}      | _        t        |j                  |j                        | _        t#        |      | _        d| _        d| j(                  j*                  v | _        | j/                          y c c}w )Nr   )r)   Fr   )r'   r(   pad_token_idpadding_idx
vocab_sizer   	Embeddingr*   embed_tokens
ModuleListrangenum_hidden_layersr   layersr   r   normr   
rotary_embgradient_checkpointingr)   r   has_sliding_layers	post_initr   s      r5   r(   zQwen2Model.__init__2  s     !.. ++LL):):F<N<NPTP`P`ammCHIaIaCbcivy1c
 !!3!39L9LM	.f=&+#"59P9P"P 	 ds   D	input_idsre   rQ   r   inputs_embedsr   r   rh   rX   c                    |d u |d uz  rt        d      || j                  |      }|r|
t               }|F||j                         nd}	t	        j
                  |	|	|j                  d   z   |j                        }||j                  d      }t        |x}
t              s:| j                  |||||d}dt        d
i |i}
| j                  rt        d
i ||
d<   |}| j                  ||      }| j                   d | j                  j"                   D ]  } ||f|
|j$                     |||||d|}! | j'                  |      }t)        ||r|	      S d 	      S )Nz:You must specify exactly one of input_ids or inputs_embedsr   r   )r   )r)   input_embedsre   r   r   rQ   full_attentionr   )re   rQ   r   r   r   r   )last_hidden_stater   r   )
ValueErrorr   r
   get_seq_lengthrF   arangerE   r   rL   r   r   r)   r   r  r   r  r   r   r   r   r   )r3   r  re   rQ   r   r  r   r   rh   past_seen_tokenscausal_mask_mappingmask_kwargsrV   r   decoder_layers                  r5   r:   zQwen2Model.forwardC  s    -t";<YZZ  --i8M0*nO!CRC^==?de"\\ "2]5H5H5K"KTaThThN )33A6L ?-F ++ -"0"0#2 ,K !"4"C{"C# &&;\;k_j;k#$78% #oom\J![[)H4;;+H+HI 
	M)	2=3O3OP).#-$7	 	M
	 		-0&+/8O
 	
>B
 	
r6   )NNNNNNN)r;   r<   r=   r    r(   r   r   r   rF   r   r   r	   FloatTensorr   r   r   r   r:   r>   r?   s   @r5   r   r   0  s    { "  151537+/59$(59E
E,,-E
 !.E
 u//0	E

 "%E
   1 12E
 D>E
 !!1!12E
 +,E
 
!E
  E
r6   r   c                   p    e Zd ZdgZddiZddgdgfiZ fdZd Zd Ze	e
	 	 	 	 	 	 	 	 	 dd	eej                     d
eej                     deej                     dee   deej                      deej                     dee   deej                     deeej                  f   dee   defd              Z xZS )Qwen2ForCausalLMzlm_head.weightlm_headcolwise_reprV   logitsc                     t         |   |       t        |      | _        |j                  | _        t        j                  |j                  |j                  d      | _        | j                          y r$   )
r'   r(   r   r   r   r   r,   r*   r  r  r2   s     r5   r(   zQwen2ForCausalLM.__init__  sU     '
 ++yy!3!3V5F5FUS 	r6   c                     || _         y r8   r   )r3   decoders     r5   set_decoderzQwen2ForCausalLM.set_decoder  s	    
r6   c                     | j                   S r8   r  r   s    r5   get_decoderzQwen2ForCausalLM.get_decoder  s    zzr6   r  re   rQ   r   r  labelsr   r   logits_to_keeprh   rX   c
                 z    | j                   d|||||||d|
}|j                  }t        |	t              rt	        |	 d      n|	}| j                  |dd|ddf         }d}|* | j                  d||| j                  j                  d|
}t        |||j                  |j                  |j                        S )a  
        Example:

        ```python
        >>> from transformers import AutoTokenizer, Qwen2ForCausalLM

        >>> model = Qwen2ForCausalLM.from_pretrained("meta-qwen2/Qwen2-2-7b-hf")
        >>> tokenizer = AutoTokenizer.from_pretrained("meta-qwen2/Qwen2-2-7b-hf")

        >>> prompt = "Hey, are you conscious? Can you talk to me?"
        >>> inputs = tokenizer(prompt, return_tensors="pt")

        >>> # Generate
        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
        ```)r  re   rQ   r   r  r   r   N)r  r  r   )lossr  r   rV   r   r   )r   r
  r   r   slicer  loss_functionr)   r   r   r   rV   r   )r3   r  re   rQ   r   r  r  r   r   r   rh   outputsrV   slice_indicesr  r"  s                   r5   r:   zQwen2ForCausalLM.forward  s    @ ,64:: 	,
)%+')	,
 	,
  118B>SV8W~ot4]kmA}a,?@A%4%%pVFt{{OeOepiopD%#33!//))
 	
r6   )	NNNNNNNNr   )r;   r<   r=   _tied_weights_keys_tp_plan_pp_planr(   r  r  r   r   r   rF   r   r   r	   r  r   r   r   r   r   r   r:   r>   r?   s   @r5   r  r    s:   *+=)H_-z:;H  151537+/59-1$(59348
E,,-8
 !.8
 u//0	8

 "%8
   1 128
 ))*8
 D>8
 !!1!128
 c5<</08
 +,8
 
 8
  8
r6   r  c                       e Zd Zy)Qwen2ForSequenceClassificationNr;   r<   r=   r   r6   r5   r+  r+        r6   r+  c                       e Zd Zy)Qwen2ForTokenClassificationNr,  r   r6   r5   r/  r/    r-  r6   r/  c                       e Zd ZdZy)Qwen2ForQuestionAnsweringtransformerN)r;   r<   r=   r   r   r6   r5   r1  r1    s    %r6   r1  )r   r   r  r+  r/  r1  )Nr   )r   )@typingr   r   r   rF   r   activationsr   cache_utilsr	   r
   
generationr   integrationsr   masking_utilsr   r   modeling_flash_attention_utilsr   modeling_layersr   r   r   r   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.genericr   configuration_qwen2r    Moduler"   rJ   rU   r   r   r`   r   r{   r}   r   r   r   r   r   r  r+  r/  r1  __all__r   r6   r5   <module>rD     s   - ,   ! . ) 7 R B  P K F & I I / ,ryy  (6	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 U\\*% % % '(%4<)RYY <)~ Y'J299 J (J(+2 +\ ?  $<299 <D Y
% Y
 Y
x N
+_ N
 N
b	%EG[ 		"?AU 	& ;=Q &r6   