a
    hR                  
   @   sX  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	 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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% ddl&m'Z' ddl(m)Z) ddl*m+Z+ G dd dej,Z-edG dd dej,Z.G dd dej,Z/dd Z0d:ddZ1ej2e3ej2d d!d"Z4d;ej,ej2ej2ej2eej2 e5e5e"e$ d$d%d&Z6G d'd( d(ej,Z7G d)d* d*eZ8eG d+d, d,e Z9eG d-d. d.e9Z:ed/d0G d1d2 d2e9eZ;ed/d0G d3d4 d4ee9Z<ed/d0G d5d6 d6ee9Z=ed/d0G d7d8 d8ee9Z>g d9Z?dS )<    )CallableOptionalUnionN)nn)auto_docstring   )ACT2FN)CacheDynamicCache)GenerationMixin)use_kernel_forward_from_hub)create_causal_mask)GenericForQuestionAnswering GenericForSequenceClassificationGenericForTokenClassificationGradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)TransformersKwargscan_return_tuple)deprecate_kwarg)check_model_inputs   )ArceeConfigc                       s$   e Zd Z fddZdd Z  ZS )ArceeMLPc                    s`   t    || _|j| _|j| _tj| j| j|jd| _tj| j| j|jd| _	t
|j | _d S )NZbias)super__init__confighidden_sizeZintermediate_sizer   LinearZmlp_biasup_proj	down_projr   Z
hidden_actact_fnselfr#   	__class__ d/var/www/html/assistant/venv/lib/python3.9/site-packages/transformers/models/arcee/modeling_arcee.pyr"   3   s    
zArceeMLP.__init__c                 C   s   |  | | |S )N)r'   r(   r&   )r*   xr-   r-   r.   forward<   s    zArceeMLP.forward)__name__
__module____qualname__r"   r0   __classcell__r-   r-   r+   r.   r   2   s   	r   ZRMSNormc                       s.   e Zd Zd fdd	Zdd Zdd Z  ZS )	ArceeRMSNormư>c                    s&   t    tt|| _|| _dS )z;
        ArceeRMSNorm is equivalent to T5LayerNorm
        N)r!   r"   r   	ParametertorchZonesweightvariance_epsilon)r*   r$   epsr+   r-   r.   r"   B   s    
zArceeRMSNorm.__init__c                 C   sJ   |j }|tj}|djddd}|t|| j  }| j|| S )N   T)Zkeepdim)	dtypetor8   float32powmeanZrsqrtr:   r9   )r*   hidden_statesZinput_dtypeZvariancer-   r-   r.   r0   J   s
    zArceeRMSNorm.forwardc                 C   s   t | jj d| j S )Nz, eps=)tupler9   shaper:   )r*   r-   r-   r.   
extra_reprQ   s    zArceeRMSNorm.extra_repr)r6   )r1   r2   r3   r"   r0   rF   r4   r-   r-   r+   r.   r5   @   s   r5   c                       sD   e Zd ZU ejed< ded fddZe e	dd Z
  ZS )	ArceeRotaryEmbeddinginv_freqNr#   c                    s   t    t|dr:t|jtr:|jd|jd| _nd| _|j| _	|j| _
|| _t| j | _| | j|\}| _| jd|dd | j| _d S )Nrope_scaling	rope_typetypedefaultrH   F)
persistent)r!   r"   hasattr
isinstancerJ   dictgetrK   Zmax_position_embeddingsZmax_seq_len_cachedZoriginal_max_seq_lenr#   r   Zrope_init_fnattention_scalingZregister_bufferrH   Zoriginal_inv_freq)r*   r#   devicerH   r+   r-   r.   r"   X   s    
zArceeRotaryEmbedding.__init__c           
      C   s   | j d d d d f  |jd dd|j}|d d d d d f  }t|jjtrl|jjdkrl|jjnd}t	j
|ddV | |  dd}t	j||fdd	}| | j }| | j }	W d    n1 s0    Y  |j|jd
|	j|jd
fS )Nr   r=   r   ZmpscpuF)device_typeZenabledr<   dim)r>   )rH   floatexpandrE   r?   rT   rP   rL   strr8   Zautocast	transposecatcosrS   sinr>   )
r*   r/   position_idsZinv_freq_expandedZposition_ids_expandedrV   ZfreqsZembr^   r_   r-   r-   r.   r0   i   s    0&,zArceeRotaryEmbedding.forward)N)r1   r2   r3   r8   Tensor__annotations__r   r"   Zno_gradr   r0   r4   r-   r-   r+   r.   rG   U   s
   

rG   c                 C   sH   | dd| j d d f }| d| j d d df }tj| |fddS )z*Rotates half the hidden dims of the input..Nr=   r<   rW   )rE   r8   r]   )r/   x1Zx2r-   r-   r.   rotate_halfy   s    rd   c                 C   sD   | |}| |}| | t| |  }|| t||  }||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.
    )	unsqueezerd   )qkr^   r_   r`   Zunsqueeze_dimZq_embedZk_embedr-   r-   r.   apply_rotary_pos_emb   s
    

rh   )rC   n_repreturnc                 C   s^   | j \}}}}|dkr| S | dddddddddf |||||} | ||| ||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   rZ   reshape)rC   ri   batchnum_key_value_headsslenhead_dimr-   r-   r.   	repeat_kv   s
    0rp           )modulequerykeyvalueattention_maskscalingdropoutkwargsc                 K   s   t || j}t || j}	t||dd| }
|d urf|d d d d d d d |jd f }|
| }
tjj|
dtj	d
|j}
tjj|
|| jd}
t|
|	}|dd }||
fS )Nr<   r   r=   )rX   r>   )ptrainingr   )rp   num_key_value_groupsr8   matmulr\   rE   r   Z
functionalZsoftmaxr@   r?   r>   rx   r|   
contiguous)rr   rs   rt   ru   rv   rw   rx   ry   
key_statesvalue_statesattn_weightscausal_maskattn_outputr-   r-   r.   eager_attention_forward   s    
&r   c                       s   e Zd ZdZeed fddZedddddej	e
ej	ej	f eej	 ee eej ee e
ej	ej	f d
ddZ  ZS )ArceeAttentionz=Multi-headed attention from 'Attention Is All You Need' paperr#   	layer_idxc                    s   t    || _|| _t|d|j|j | _|j|j | _	| jd | _
|j| _d| _tj|j|j| j |jd| _tj|j|j| j |jd| _tj|j|j| j |jd| _tj|j| j |j|jd| _d S )Nro   g      Tr    )r!   r"   r#   r   getattrr$   Znum_attention_headsro   rm   r}   rw   attention_dropoutZ	is_causalr   r%   Zattention_biasq_projk_projv_projo_projr*   r#   r   r+   r-   r.   r"      s(    
zArceeAttention.__init__past_key_valuepast_key_values4.58new_nameversionN)rC   position_embeddingsrv   r   cache_positionry   rj   c                 K   s$  |j d d }g |d| jR }| ||dd}	| ||dd}
| ||dd}|\}}t|	|
||\}	}
|d ur|||d}||
|| j	|\}
}t
}| jjdkrt| jj }|| |	|
||f| jsdn| j| jd|\}}|jg |dR   }| |}||fS )Nr=   r   r<   )r_   r^   r   eagerrq   )rx   rw   )rE   ro   r   viewr\   r   r   rh   updater   r   r#   Z_attn_implementationr   r|   r   rw   rk   r   r   )r*   rC   r   rv   r   r   ry   Zinput_shapeZhidden_shapeZquery_statesr   r   r^   r_   Zcache_kwargsZattention_interfacer   r   r-   r-   r.   r0      s8    


zArceeAttention.forward)NN)r1   r2   r3   __doc__r   intr"   r   r8   ra   rD   r   r	   
LongTensorr   r   r0   r4   r-   r-   r+   r.   r      s     r   c                       s   e Zd Zeed fddZedddddeje	ej e	ej
 e	e e	e e	ej
 e	eejejf  ee ejd
	ddZ  ZS )ArceeDecoderLayerr   c                    sR   t    |j| _t||d| _t|| _t|j|jd| _	t|j|jd| _
d S )Nr   r;   )r!   r"   r$   r   	self_attnr   mlpr5   rms_norm_epsinput_layernormpost_attention_layernormr   r+   r-   r.   r"   	  s    

zArceeDecoderLayer.__init__r   r   r   r   NF)	rC   rv   r`   r   	use_cacher   r   ry   rj   c              
   K   s^   |}	|  |}| jf |||||||d|\}}
|	| }|}	| |}| |}|	| }|S )N)rC   rv   r`   r   r   r   r   )r   r   r   r   )r*   rC   rv   r`   r   r   r   r   ry   Zresidual_r-   r-   r.   r0     s&    




zArceeDecoderLayer.forward)NNNFNN)r1   r2   r3   r   r   r"   r   r8   ra   r   r   r	   boolrD   r   r   r0   r4   r-   r-   r+   r.   r     s&   
      r   c                   @   sH   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dS )ArceePreTrainedModelr#   modelTr   r   )rC   
attentionsN)r1   r2   r3   r   rb   base_model_prefixZsupports_gradient_checkpointingZ_no_split_modulesZ_skip_keys_device_placementZ_supports_flash_attnZ_supports_sdpaZ_supports_flex_attnZ_can_compile_fullgraphZ_supports_attention_backendr   r   Z_can_record_outputsr-   r-   r-   r.   r   6  s   
r   c                       st   e Zd Zed fddZeedeej	 eej
 eej	 ee eej eej	 ee ee ed	ddZ  ZS )	
ArceeModelrI   c                    s   t     j| _ j| _t j j| j| _t	 fddt
 jD | _t j jd| _t d| _d| _|   d S )Nc                    s   g | ]}t  |qS r-   )r   ).0r   rI   r-   r.   
<listcomp>R      z'ArceeModel.__init__.<locals>.<listcomp>r   rI   F)r!   r"   Zpad_token_idZpadding_idx
vocab_sizer   Z	Embeddingr$   embed_tokensZ
ModuleListrangenum_hidden_layerslayersr5   r   normrG   
rotary_embZgradient_checkpointing	post_initr)   r+   rI   r.   r"   K  s    zArceeModel.__init__N)		input_idsrv   r`   r   inputs_embedsr   r   ry   rj   c              	   K   s   |d u |d uA rt d|d u r*| |}|rB|d u rBt| jd}|d u rz|d urZ| nd}	tj|	|	|jd  |jd}|d u r|	d}t
| j|||||d}
|}| ||}| jd | jj D ] }||f|
||||d|}q| |}t||dS )	Nz:You must specify exactly one of input_ids or inputs_embedsrI   r   r   )rT   )r#   Zinput_embedsrv   r   r   r`   )rv   r`   r   r   r   )last_hidden_stater   )
ValueErrorr   r
   r#   Zget_seq_lengthr8   ZarangerE   rT   re   r   r   r   r   r   r   )r*   r   rv   r`   r   r   r   r   ry   Zpast_seen_tokensr   rC   r   Zdecoder_layerr-   r-   r.   r0   [  sP    

	

zArceeModel.forward)NNNNNNN)r1   r2   r3   r   r"   r   r   r   r8   r   ra   r	   FloatTensorr   r   r   r   r0   r4   r-   r-   r+   r.   r   I  s*          r   zarcee-ai/AFM-4.5B)
checkpointc                       s   e Zd ZdgZddiZddgdgfiZ fddZeede	e
j e	e
j e	e
j e	e e	e
j e	e
j e	e e	e
j eee
jf ee ed
ddZ  ZS )ArceeForCausalLMzlm_head.weightlm_headZcolwise_reprC   logitsc                    s@   t  | t|| _|j| _tj|j|jdd| _| 	  d S )NFr    )
r!   r"   r   r   r   r   r%   r$   r   r   r)   r+   r-   r.   r"     s
    
zArceeForCausalLM.__init__Nr   )r   rv   r`   r   r   labelsr   r   logits_to_keepry   rj   c
              
   K   s   | j f |||||||d|
}|j}t|	tr<t|	 dn|	}| |dd|ddf }d}|dur| jf ||| jjd|
}t	|||j
|j|jdS )a  
        Example:

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

        >>> model = ArceeForCausalLM.from_pretrained("meta-arcee/Arcee-2-7b-hf")
        >>> tokenizer = AutoTokenizer.from_pretrained("meta-arcee/Arcee-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   rv   r`   r   r   r   r   N)r   r   r   )lossr   r   rC   r   )r   r   rP   r   slicer   Zloss_functionr#   r   r   r   rC   r   )r*   r   rv   r`   r   r   r   r   r   r   ry   outputsrC   Zslice_indicesr   r   r-   r-   r.   r0     s0     zArceeForCausalLM.forward)	NNNNNNNNr   )r1   r2   r3   Z_tied_weights_keysZ_tp_planZ_pp_planr"   r   r   r   r8   r   ra   r	   r   r   r   r   r   r   r   r0   r4   r-   r-   r+   r.   r     s8   	         r   c                   @   s   e Zd ZdS )ArceeForSequenceClassificationNr1   r2   r3   r-   r-   r-   r.   r     s   r   c                   @   s   e Zd ZdZdS )ArceeForQuestionAnsweringZtransformerN)r1   r2   r3   r   r-   r-   r-   r.   r     s   r   c                   @   s   e Zd ZdS )ArceeForTokenClassificationNr   r-   r-   r-   r.   r     s   r   )r   r   r   r   r   r   )Nr   )rq   )@typingr   r   r   r8   r   Ztransformers.utilsr   Zactivationsr   Zcache_utilsr	   r
   Z
generationr   Zintegrationsr   Zmasking_utilsr   Zmodeling_layersr   r   r   r   Zmodeling_outputsr   r   Zmodeling_rope_utilsr   r   Zmodeling_utilsr   r   Zprocessing_utilsr   utilsr   r   Zutils.deprecationr   Zutils.genericr   Zconfiguration_arceer   Moduler   r5   rG   rd   rh   ra   r   rp   rY   r   r   r   r   r   r   r   r   r   __all__r-   r-   r-   r.   <module>   sd   $
 G.NK