
    qi=                        d Z ddlmZ ddl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 dd
l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mZmZ ddl m!Z!  ejD                  e#      Z$ G d dejJ                        Z& G d de      Z'd Z(d$dZ) G d de      Z* G d de      Z+ G d de
      Z, G d d e      Z- G d! d"e      Z.g d#Z/y)%zPyTorch Cohere model.    )CallableN)nn   )Cache)FlashAttentionKwargs)GradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)dynamic_rope_update)ALL_ATTENTION_FUNCTIONS)Unpack)TransformersKwargslogging)maybe_autocast   )LlamaAttentionLlamaForCausalLMLlamaMLP
LlamaModelLlamaRotaryEmbeddingeager_attention_forward   )CohereConfigc                   &     e Zd Zd fd	Zd Z xZS )CohereLayerNormc                     t         |           t        j                  t	        j
                  |            | _        || _        y)zcThe hidden size can be a tuple or an int. The tuple is used for QKNorm to normalize across head_dimN)super__init__r   	Parametertorchonesweightvariance_epsilon)selfhidden_sizeepsbias	__class__s       [/opt/pipecat/venv/lib/python3.12/site-packages/transformers/models/cohere/modular_cohere.pyr   zCohereLayerNorm.__init__5   s/    ll5::k#:; #    c                    |j                   }|j                  t        j                        }|j	                  dd      }||z
  j                  d      j	                  dd      }||z
  t        j                  || j                  z         z  }| j                  j                  t        j                        |z  }|j                  |      S )NT)keepdimr   )	dtypetor    float32meanpowrsqrtr#   r"   )r$   hidden_statesinput_dtyper1   variances        r)   forwardzCohereLayerNorm.forward;   s    #))%((7!!"d!3!D(--a055b$5G&-XH]H]=]1^^u}}5E,,r*   )Ngh㈵>F)__name__
__module____qualname__r   r7   __classcell__r(   s   @r)   r   r   4   s    $-r*   r   c                   D    e Zd Z ej                         ed               Zy)CohereRotaryEmbeddingc                    | j                   d d d d f   j                         j                  |j                  d   dd      }|d d d d d f   j                         }t	        |j
                  j                  t              r/|j
                  j                  dk7  r|j
                  j                  nd}t        |d      5  |j                         |j                         z  j                  dd      }t        j                  |dd	      }|j                         | j                  z  }|j                         | j                  z  }	d d d        j                  |j                   
      	j                  |j                   
      fS # 1 sw Y   AxY w)Nr   r,   r   mpscpuF)device_typeenabledr   dimr.   )inv_freqfloatexpandshape
isinstancedevicetypestrr   	transposer    repeat_interleavecosattention_scalingsinr/   r.   )
r$   xposition_idsinv_freq_expandedposition_ids_expandedrB   freqsembrQ   rS   s
             r)   r7   zCohereRotaryEmbedding.forwardF   s@    !MM$4-8>>@GGHZHZ[\H]_acde ,QaZ 8 > > @'1!((--'E!((--[`J`ahhmmfkUC 	5&,,.1F1L1L1NNYYZ[]^_E))%;C'')d444C'')d444C		5 vvAGGv$cff177f&;;;	5 	5s   =BFF
N)r8   r9   r:   r    no_gradr   r7    r*   r)   r>   r>   E   s$    U]]_<  <r*   r>   c                     | dd d df   }| ddd df   }t        j                  | |gd      j                  d      }|S )N.r   r   r,   rD   )r    stackflatten)rT   x1x2rot_xs       r)   rotate_halfrc   V   sL    	
3!8B	
319BKK"b	r*2226ELr*   c                 6   | j                   }| j                         } |j                         }|j                  |      }|j                  |      }| |z  t        |       |z  z   }||z  t        |      |z  z   }|j	                  |      |j	                  |      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.
        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.
    rF   )r.   rH   	unsqueezerc   r/   )qkrQ   rS   unsqueeze_dimr.   q_embedk_embeds           r)   apply_rotary_pos_embrk   ^   s    $ GGE		A		A
--
&C
--
&C3w;q>C/0G3w;q>C/0G::E:"GJJUJ$;;;r*   c                        e Zd Z fdZ xZS )	CohereMLPc                 J   t         |   |       t        j                  | j                  | j
                  d      | _        t        j                  | j                  | j
                  d      | _        t        j                  | j
                  | j                  d      | _        y )NF)r'   )	r   r   r   Linearr%   intermediate_size	gate_projup_proj	down_projr$   configr(   s     r)   r   zCohereMLP.__init__{   ss     4#3#3T5K5KRWXyy!1!143I3IPUV4#9#94;K;KRWXr*   )r8   r9   r:   r   r;   r<   s   @r)   rm   rm   z   s    Y Yr*   rm   c                       e Zd ZdZddededz  f fdZ	 	 ddej                  de	ej                  ej                  f   dej                  dz  d	e
dz  d
ej                  dz  dee   de	ej                  ej                  dz  f   fdZ xZS )CohereAttentionz=Multi-headed attention from 'Attention Is All You Need' paperNru   	layer_idxc                 *   t         |   ||       |j                  | _        | j                  ret        |j                  | j
                  f|j                        | _        t        |j                  | j
                  f|j                        | _	        y y Nr%   r&   )
r   r   use_qk_normr   num_attention_headshead_dimlayer_norm_epsq_normnum_key_value_headsk_normr$   ru   rx   r(   s      r)   r   zCohereAttention.__init__   s|    +!--)#77GVMbMbDK *#77GVMbMbDK r*   r4   position_embeddingsattention_maskpast_key_valuescache_positionkwargsreturnc                    |j                   d d }g |d| j                  }| j                  |      j                  |      }	| j	                  |      j                  |      }
| j                  |      j                  |      }| j                  r"| j                  |	      }	| j                  |
      }
|	j                  dd      }	|
j                  dd      }
|j                  dd      }|\  }}t        |	|
||      \  }	}
|'|||d}|j                  |
|| j                  |      \  }
}t        j                  | j                  j                   t"              } || |	|
||f| j$                  sdn| j&                  | j(                  d|\  }} |j*                  g |d j-                         }| j/                  |      }||fS )Nr,   r   r   )rS   rQ   r   g        )dropoutscaling)rJ   r~   q_projviewk_projv_projr|   r   r   rO   rk   updaterx   r   get_interfaceru   _attn_implementationr   trainingattention_dropoutr   reshape
contiguouso_proj)r$   r4   r   r   r   r   r   input_shapehidden_shapequery_states
key_statesvalue_statesrQ   rS   cache_kwargsattention_interfaceattn_outputattn_weightss                     r)   r7   zCohereAttention.forward   s    $))#2.88b8$--8{{=166|D[[/44\B
{{=166|D;;|4LZ0J#--a3))!Q/
#--a3&S#7jRUWZ#[ j&#&snUL'6'='=j,X\XfXfht'u$J(?(M(MKK,,.E)
 %8	%
  $}}C$2H2HLL	%
 	%
!\ *k));;;;FFHkk+.L((r*   N)NN)r8   r9   r:   __doc__r   intr   r    Tensortupler   
LongTensorr   r   r7   r;   r<   s   @r)   rw   rw      s    G
| 
d
 
" )-261)||1) #5<<#=>1) t+	1)
 1) ((4/1) -.1) 
u||U\\D00	11)r*   rw   c                   d    e Zd Zdedef fdZ	 	 	 	 	 	 ddej                  dej                  dz  dej                  dz  de	dz  d	e
dz  d
ej                  dz  deej                  ej                  f   dz  dee   deej                  eej                  ej                  f   dz  f   fdZ xZS )CohereDecoderLayerru   rx   c                     t         |           |j                  | _        t        ||      | _        t        |      | _        t        |j                  |j                        | _	        y )N)ru   rx   r{   )
r   r   r%   rw   	self_attnrm   mlpr   r   input_layernormr   s      r)   r   zCohereDecoderLayer.__init__   sR    !--()LV$.F<N<NU[UjUjkr*   Nr4   r   rU   r   	use_cacher   r   r   r   c                     |}	| j                  |      } | j                  d|||||||d|\  }
}| j                  |      }|	|
z   |z   }|S )ar  
        Args:
            hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
            attention_mask (`torch.FloatTensor`, *optional*):
                attention mask of size `(batch_size, sequence_length)` if flash attention is used or `(batch_size, 1,
                query_sequence_length, key_sequence_length)` if default attention is used.
            past_key_values (`Cache`, *optional*): cached past key and value projection states
            output_attentions (`bool`, *optional*):
                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
                returned tensors for more detail.
            use_cache (`bool`, *optional*):
                If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
                (see `past_key_values`).
            cache_position (`torch.LongTensor` of shape `(sequence_length)`, *optional*):
                Indices depicting the position of the input sequence tokens in the sequence
            position_embeddings (`tuple[torch.FloatTensor, torch.FloatTensor]`, *optional*):
                Tuple containing the cosine and sine positional embeddings of shape `(batch_size, seq_len, head_dim)`,
                with `head_dim` being the embedding dimension of each attention head.
        )r4   r   rU   r   r   r   r   r[   )r   r   r   )r$   r4   r   rU   r   r   r   r   r   residualhidden_states_attention_hidden_states_mlps                r)   r7   zCohereDecoderLayer.forward   s{    < !,,];%3T^^ 	&
')%+) 3	&
 	&
" !HH]3 #::=NNr*   )NNNFNN)r8   r9   r:   r   r   r   r    r   r   r   boolr   r   r   FloatTensorr7   r;   r<   s   @r)   r   r      s    l| l l /304(,!&26HL.||. t+. &&-	.
 . $;. ((4/. #5<<#=>E. -.. 
u  %(9(95;L;L(L"MPT"TT	U.r*   r   c                   $     e Zd Zdef fdZ xZS )CohereModelru   c           	         t         |   |       t        j                  t	        |j
                        D cg c]  }t        ||       c}      | _        t        |j                  |j                        | _        y c c}w rz   )r   r   r   
ModuleListrangenum_hidden_layersr   layersr   r%   r   normr   s      r)   r   zCohereModel.__init__   sc     mmDI&JbJbDcdy	2d
 $1C1C&J_J_`	 es   A=)r8   r9   r:   r   r   r;   r<   s   @r)   r   r      s    a| a ar*   r   c                   F    e Zd Z fdZ	 	 	 	 	 	 	 	 	 	 	 ddej
                  dz  dej                  dz  dej
                  dz  dedz  dej                  dz  dej
                  dz  d	e	dz  d
e	dz  de	dz  dej
                  dz  de
ej                  z  dee   defdZ xZS )CohereForCausalLMc                     t         |   |       t        |      | _        |j                  | _        |j
                  | _        y r   )r   r   r   modellogit_scaletie_word_embeddingsrt   s     r)   r   zCohereForCausalLM.__init__  s8      (
!--#)#=#= r*   N	input_idsr   rU   r   inputs_embedslabelsr   output_attentionsoutput_hidden_statesr   logits_to_keepr   r   c                    ||n| j                   j                  }|	|	n| j                   j                  }	 | j                  d||||||||	|
d	|}|j                  }t        |t              rt        | d      n|}| j                  |dd|ddf         }|| j                  z  }d}|* | j                  d||| j                   j                  d|}t        |||j                  |j                  |j                        S )az  
        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, CohereForCausalLM

        >> model = CohereForCausalLM.from_pretrained("CohereForAI/c4ai-command-r-v01")
        >> tokenizer = AutoTokenizer.from_pretrained("CohereForAI/c4ai-command-r-v01")

        >> 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."
        ```N)	r   r   rU   r   r   r   r   r   r   )logitsr   
vocab_size)lossr   r   r4   
attentionsr[   )ru   r   r   r   last_hidden_staterK   r   slicelm_headr   loss_functionr   r
   r   r4   r   )r$   r   r   rU   r   r   r   r   r   r   r   r   r   outputsr4   slice_indicesr   r   s                     r)   r7   zCohereForCausalLM.forward  s+   J 2C1N-TXT_T_TqTq$8$D $++JjJj 	
 ,64:: ,
)%+'/!5),
 ,
  118B>SV8W~ot4]kmA}a,?@A$***%4%%pVFt{{OeOepiopD%#33!//))
 	
r*   )NNNNNNNNNNr   )r8   r9   r:   r   r    r   r   r   r   r   r   r   r   r
   r7   r;   r<   s   @r)   r   r     s   > .2.204(,26*.!%)-,026-.H
##d*H
 t+H
 &&-	H

 H
 ((4/H
   4'H
 $;H
  $;H
 #TkH
 ((4/H
 ell*H
 +,H
 
 H
r*   r   )r   r   CoherePreTrainedModel)r   )0r   collections.abcr   r    r   cache_utilsr   modeling_flash_attention_utilsr   modeling_layersr   modeling_outputsr	   r
   modeling_rope_utilsr   modeling_utilsr   processing_utilsr   utilsr   r   utils.genericr   llama.modeling_llamar   r   r   r   r   r   configuration_coherer   
get_loggerr8   loggerModuler   r>   rc   rk   rm   rw   r   r   r   __all__r[   r*   r)   <module>r      s   ,  $     B 9 O 6 5 & 0 +  / 
		H	%-bii -"<0 <"<8Y Y@)n @)F63 6ra* aO
( O
dr*   