
    rhBV                     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 dd	lmZ dd
l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jV                        Z, G d de      Z-dej\                  de/dej\                  fdZ0	 d8dejV                  dej\                  dej\                  dej\                  deej\                     d e1d!e1d"e"e$   fd#Z2d$ Z3d9d%Z4 G d& d'ejV                        Z5 ed(       G d) d*ejV                               Z6 G d+ d,ejV                        Z7e% G d- d.e              Z8e% G d/ d0e8             Z9e% G d1 d2e8e             Z: G d3 d4ee8      Z; G d5 d6ee8      Z<g d7Z=y):    )CallableOptionalUnionN   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hub)create_causal_mask)FlashAttentionKwargs) 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   )
Glm4Configc                   V     e Zd Z fdZdej
                  dej
                  fdZ xZS )Glm4MLPc                 *   t         |           || _        t        j                  |j
                  d|j                  z  d      | _        t        j                  |j                  |j
                  d      | _        t        |j                     | _        y )N   Fbias)super__init__confignnLinearhidden_sizeintermediate_sizegate_up_proj	down_projr   
hidden_actactivation_fnselfr&   	__class__s     y/var/www/html/ai-insurance-compliance-backend/venv/lib/python3.12/site-packages/transformers/models/glm4/modeling_glm4.pyr%   zGlm4MLP.__init__0   sp    IIf&8&8!f>V>V:V]bc6#;#;V=O=OV[\#F$5$56    hidden_statesreturnc                     | j                  |      }|j                  dd      \  }}|| j                  |      z  }| j                  |      S )Nr!   dim)r+   chunkr.   r,   )r0   r4   	up_statesgates       r2   forwardzGlm4MLP.forward8   sL    %%m4	#//!/4i 2 24 88	~~i((r3   )__name__
__module____qualname__r%   torchFloatTensorr=   __classcell__r1   s   @r2   r   r   /   s'    7)U%6%6 )5;L;L )r3   r   c                   d    e Zd Zdedef fdZ	 	 	 	 	 	 ddej                  deej                     deej                     dee
   dee   d	eej                     d
eeej                  ej                  f      dee   deej                  eeej                  ej                  f      f   fdZ xZS )Glm4DecoderLayerr&   	layer_idxc                    t         |           |j                  | _        t        ||      | _        t        |      | _        t        |j                  |j                        | _	        t        |j                  |j                        | _
        t        |j                  |j                        | _        t        |j                  |j                        | _        y )N)r&   rG   eps)r$   r%   r)   Glm4Attention	self_attnr   mlpGlm4RMSNormrms_norm_epsinput_layernormpost_attention_layernormpost_self_attn_layernormpost_mlp_layernormr0   r&   rG   r1   s      r2   r%   zGlm4DecoderLayer.__init__B   s    !--&f	J6?*6+=+=6CVCVW(3F4F4FFL_L_(`%(3F4F4FFL_L_(`%"-f.@.@fFYFY"Zr3   r4   attention_maskposition_idspast_key_value	use_cachecache_positionposition_embeddingskwargsr5   c                    |}	| j                  |      } | j                  d|||||||d|\  }}
| j                  |      }|	|z   }|}	| j                  |      }| j	                  |      }| j                  |      }|	|z   }|S )N)r4   rU   rV   rW   rX   rY   rZ    )rP   rL   rR   rQ   rM   rS   )r0   r4   rU   rV   rW   rX   rY   rZ   r[   residual_s              r2   r=   zGlm4DecoderLayer.forwardM   s     !,,];)4>> 	
')%)) 3	
 	
q 55mD =0 55mD///> =0r3   )NNNFNN)r>   r?   r@   r   intr%   rA   Tensorr   
LongTensorr   booltupler   r   rB   r=   rC   rD   s   @r2   rF   rF   A   s    	[z 	[c 	[ 2637*.$)59KO!||! !.! u//0	!
 !! D>! !!1!12! &eELL%,,,F&GH! -.! 
u  (51B1BEDUDU1U+V"WW	X!r3   rF   r4   n_repr5   c                     | j                   \  }}}}|dk(  r| S | dddddddddf   j                  |||||      } | j                  |||z  ||      S )z
    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
    r   N)shapeexpandreshape)r4   re   batchnum_key_value_headsslenhead_dims         r2   	repeat_kvrn   q   so    
 2?1D1D.Ehz!!Qa"23::5BUW\^bdlmM  (;e(CT8TTr3   modulequerykeyvaluerU   scalingdropoutr[   c                 T   t        || j                        }t        || j                        }	t        j                  ||j	                  dd            |z  }
|#|d d d d d d d |j
                  d   f   }|
|z   }
t        j                  j                  |
dt        j                        j                  |j                        }
t        j                  j                  |
|| j                        }
t        j                  |
|	      }|j	                  dd      j                         }||
fS )Nr!   r   r7   )r9   dtype)ptrainingr   )rn   num_key_value_groupsrA   matmul	transposerg   r'   
functionalsoftmaxfloat32torw   rt   ry   
contiguous)ro   rp   rq   rr   rU   rs   rt   r[   
key_statesvalue_statesattn_weightscausal_maskattn_outputs                r2   eager_attention_forwardr   }   s    3 ; ;<JUF$?$?@L<<z';';Aq'ABWLL!$Q1.D
0@0@0D.D%DE#k1==((2U]](SVVW\WbWbcL==((6??([L,,|\:K''1-88:K$$r3   c                 |    | ddddf   }| ddddf   }t        j                  | |fd      j                  d      S )	z*Rotates half the hidden dims of the input..r   Nr!   r   r7   r8   rv   )rA   stackflatten)xx1x2s      r2   rotate_halfr      sJ    	
319B	
319B;;Ryb)11"55r3   c                    |j                  |      }|j                  |      }|dd|j                  d   dz  f   j                  dd      }|dd|j                  d   dz  f   j                  dd      }|j                  d   }| dd|f   | d|df   }}|dd|f   |d|df   }
}	||z  t        |      |z  z   }|	|z  t        |	      |z  z   }t	        j
                  ||gd      }t	        j
                  ||
gd      }||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.
    .Nr7   r!   r8   )	unsqueezerg   repeat_interleaver   rA   cat)qkcossinrV   unsqueeze_dim
rotary_dimq_rotq_passk_rotk_passq_embedk_embeds                r2   apply_rotary_pos_embr      sD   ( --
&C
--
&C c'SYYr]a'''
(
:
:1"
:
EC
c'SYYr]a'''
(
:
:1"
:
EC 2Jc;J;&'3
+;)<6Ec;J;&'3
+;)<6E s{{51C78Gs{{51C78G ii&)r2Gii&)r2GGr3   c                       e Zd ZdZddede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j                  f   fdZ xZS )rK   z=Multi-headed attention from 'Attention Is All You Need' paperr&   rG   c                 P   t         |           || _        || _        t	        |d|j
                  |j                  z        | _        |j                  |j                  z  | _	        | j                  dz  | _
        |j                  | _        d| _        t        j                  |j
                  |j                  | j                  z  |j                        | _        t        j                  |j
                  |j                  | j                  z  |j                        | _        t        j                  |j
                  |j                  | j                  z  |j                        | _        t        j                  |j                  | j                  z  |j
                  d      | _        y )Nrm   g      Tr"   F)r$   r%   r&   rG   getattrr)   num_attention_headsrm   rk   rz   rs   attention_dropout	is_causalr'   r(   attention_biasq_projk_projv_projo_projrT   s      r2   r%   zGlm4Attention.__init__   sD   "
F4F4F&JdJd4de$*$>$>&B\B\$\!}}d*!'!9!9ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JQWQfQf
 ii : :T]] JFL^L^ejkr3   r4   rZ   rU   rW   rY   r[   r5   c                 4   |j                   d d }g |d| j                  }| j                  |      j                  |      j	                  dd      }	| j                  |      j                  |      j	                  dd      }
| j                  |      j                  |      j	                  dd      }|\  }}t        |	|
||      \  }	}
|'|||d}|j                  |
|| j                  |      \  }
}t        }| j                  j                  dk7  rt        | j                  j                     } || |	|
||f| j                  sdn| j                  | j                   d|\  }} |j"                  g |d j%                         }| j'                  |      }||fS )Nr7   r   r!   )r   r   rY   eager        )rt   rs   )rg   rm   r   viewr|   r   r   r   updaterG   r   r&   _attn_implementationr   ry   r   rs   ri   r   r   )r0   r4   rZ   rU   rW   rY   r[   input_shapehidden_shapequery_statesr   r   r   r   cache_kwargsattention_interfacer   r   s                     r2   r=   zGlm4Attention.forward   s    $))#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$2H2HLL	%
 	%
!\ *k));;;;FFHkk+.L((r3   N)NN)r>   r?   r@   __doc__r   r   r`   r%   rA   ra   rd   r   rb   r   r   r=   rC   rD   s   @r2   rK   rK      s    Glz lhsm l4 +/59))||)) #5<<#=>)) !.	))
 !)) !!1!12)) +,)) 
u||U\\)	*))r3   rK   RMSNormc                   ,     e Zd Zd fd	Zd Zd Z xZS )rN   c                     t         |           t        j                  t	        j
                  |            | _        || _        y)z:
        Glm4RMSNorm is equivalent to T5LayerNorm
        N)r$   r%   r'   	ParameterrA   onesweightvariance_epsilon)r0   r)   rJ   r1   s      r2   r%   zGlm4RMSNorm.__init__  s1     	ll5::k#:; #r3   c                 "   |j                   }|j                  t        j                        }|j	                  d      j                  dd      }|t        j                  || j                  z         z  }| j                  |j                  |      z  S )Nr!   r7   T)keepdim)	rw   r   rA   r   powmeanrsqrtr   r   )r0   r4   input_dtypevariances       r2   r=   zGlm4RMSNorm.forward  sy    #))%((7 $$Q',,R,>%Ht?T?T4T(UU{{]--k:::r3   c                 ^    t        | j                  j                         d| j                   S )Nz, eps=)rd   r   rg   r   r0   s    r2   
extra_reprzGlm4RMSNorm.extra_repr  s*    ))*+6$2G2G1HIIr3   )gư>)r>   r?   r@   r%   r=   r   rC   rD   s   @r2   rN   rN     s    $;Jr3   rN   c                   ^     e Zd Zddef fdZ ej                         ed               Z xZ	S )Glm4RotaryEmbeddingr&   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)r0   r&   devicer   r1   s       r2   r%   zGlm4RotaryEmbedding.__init__"  s    6>*z&:M:Mt/T#0044[&BUBUBYBYZ`BabDN&DN"("@"@$*$B$B!/?+/+<+<T[[&+Q($(ZeD!%r3   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   r7   r   mpscpuF)device_typeenabledr!   r8   )rw   )r   floatrh   rg   r   r   r   r   strrA   autocastr|   r   r   r   r   rw   )
r0   r   rV   inv_freq_expandedposition_ids_expandedr   freqsembr   r   s
             r2   r=   zGlm4RotaryEmbedding.forward3  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.r   )
r>   r?   r@   r   r%   rA   no_gradr   r=   rC   rD   s   @r2   r   r   !  s3    /z /" U]]_<  <r3   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)Glm4PreTrainedModelr&   modelTrF   past_key_values)r4   
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_backendrF   rK   _can_record_outputsr]   r3   r2   r   r   C  sQ    &*#+,#4"5N!"&)#r3   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j                     d	ee   d
ee   defd              Z xZS )	Glm4Modelr&   c           	         t         |   |       |j                  | _        |j                  | _        t        j                  |j                  |j                  | j                        | _        t        j                  t        |j                        D cg c]  }t        ||       c}      | _        t        |j                  |j                        | _        t#        |      | _        d| _        | j)                          y c c}w )NrI   )r&   F)r$   r%   pad_token_idpadding_idx
vocab_sizer'   	Embeddingr)   embed_tokens
ModuleListrangenum_hidden_layersrF   layersrN   rO   normr   
rotary_embgradient_checkpointing	post_initrT   s      r2   r%   zGlm4Model.__init__X  s     !.. ++LL):):F<N<NPTP`P`ammBGH`H`BabYfi0b
   2 28K8KL	-V<&+# 	 cs   D	input_idsrU   rV   r   inputs_embedsrY   rX   r[   r5   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        | j                  |||||      }
|}| j                  ||      }| j                  d | j                  j                   D ]  } ||f|
||||d|} | j                  |      }t        ||      S )Nz:You must specify exactly one of input_ids or inputs_embedsr   r   )r   )r&   input_embedsrU   rY   r   rV   )rU   rV   rW   rY   rZ   )last_hidden_stater   )
ValueErrorr  r	   get_seq_lengthrA   arangerg   r   r   r   r&   r  r  r  r  r   )r0   r  rU   rV   r   r  rY   rX   r[   past_seen_tokensr   r4   rZ   decoder_layers                 r2   r=   zGlm4Model.forwardh  sT    -t";<YZZ *.*;*;I*FM0*nO!CRC^==?de+0<< "2]5H5H5K"KTaThTh,N )33A6L(;;&))+%
 &"oom\J![[)H4;;+H+HI 		M)*).-$7 M		 		-0&++
 	
r3   )NNNNNNN)r>   r?   r@   r   r%   r   r   r   rA   rb   ra   r   rB   rc   r   r   r   r=   rC   rD   s   @r2   r   r   V  s    z    151537+/5959$(8
E,,-8
 !.8
 u//0	8

 "%8
   1 128
 !!1!128
 D>8
 +,8
 
!8
  8
r3   r   c                   z    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eef   fd              Z xZS )Glm4ForCausalLMzlm_head.weightlm_headcolwise_repr4   logitsc                     t         |   |       t        |      | _        |j                  | _        t        j                  |j                  |j                  d      | _        | j                          y )NFr"   )
r$   r%   r   r   r   r'   r(   r)   r  r
  r/   s     r2   r%   zGlm4ForCausalLM.__init__  sU     v&
 ++yy!3!3V5F5FUS 	r3   c                     || _         y r   r   )r0   decoders     r2   set_decoderzGlm4ForCausalLM.set_decoder  s	    
r3   c                     | j                   S r   r  r   s    r2   get_decoderzGlm4ForCausalLM.get_decoder  s    zzr3   r  rU   rV   r   r  labelsrX   rY   logits_to_keepr[   r5   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 )ah  
        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
            config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.

        Example:

        ```python
        >>> from transformers import AutoTokenizer, Glm4ForCausalLM

        >>> model = Glm4ForCausalLM.from_pretrained("THUDM/GLM-4-9B-0414")
        >>> tokenizer = AutoTokenizer.from_pretrained("THUDM/GLM-4-9B-0414")

        >>> 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  rU   rV   r   r  rX   rY   N)r  r!  r   )lossr  r   r4   r   r]   )r   r  r   r`   slicer  loss_functionr&   r   r   r   r4   r   )r0   r  rU   rV   r   r  r!  rX   rY   r"  r[   outputsr4   slice_indicesr  r$  s                   r2   r=   zGlm4ForCausalLM.forward  s    J ,64:: 	,
)%+')	,
 	,
  118B>SV8W~ot4]kmA}a,?@A%4%%pVFt{{OeOepiopD%#33!//))
 	
r3   )	NNNNNNNNr   )r>   r?   r@   _tied_weights_keys_tp_plan_pp_planr%   r  r   r   r   r   rA   rb   ra   r   rB   rc   r   r`   r   r   rd   r   r=   rC   rD   s   @r2   r  r    sE   *+=)H_-z:;H  151537+/59-1$(5934=
E,,-=
 !.=
 u//0	=

 "%=
   1 12=
 ))*=
 D>=
 !!1!12=
 c5<</0=
 +,=
 
u,,	-=
  =
r3   r  c                       e Zd Zy)Glm4ForSequenceClassificationNr>   r?   r@   r]   r3   r2   r-  r-        r3   r-  c                       e Zd Zy)Glm4ForTokenClassificationNr.  r]   r3   r2   r1  r1     r/  r3   r1  )r   r   r  r-  r1  )r   )Nr   )>typingr   r   r   rA   torch.nnr'   activationsr   cache_utilsr   r	   
generationr
   integrationsr   masking_utilsr   modeling_flash_attention_utilsr   modeling_layersr   r   r   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.genericr   configuration_glm4r   Moduler   rF   ra   r`   rn   r   r   r   r   rK   rN   r   r   r   r  r-  r1  __all__r]   r3   r2   <module>rD     s  , - ,   ! . ) 7 / B 
 P K F & I I / *)bii )$-1 -`	UU\\ 	U# 	U%,, 	U& %II%<<% 
% <<	%
 U\\*% % % '(%46'TA)BII A)H Y'J")) J (J(<")) <D /  $ K
# K
 K
\ S
)? S
 S
l	$DFY 		!>@S 	r3   