o
    i!T                     @   s  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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 dej+Z,dd Z-d;ddZ.dej/de0dej/fddZ1	d<d ej+d!ej/d"ej/d#ej/d$eej/ d%e2d&e2d'e"e$ fd(d)Z3G d*d+ d+ej+Z4G d,d- d-eZ5G d.d/ d/ej+Z6e%G d0d1 d1e Z7e%G d2d3 d3e7Z8e%G d4d5 d5e7eZ9G d6d7 d7ee7Z:G d8d9 d9ee7Z;g d:Z<dS )=    )CallableOptionalUnionN)nn)check_model_inputs   )ACT2FN)CacheDynamicCache)GenerationMixin)create_causal_mask!create_sliding_window_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)deprecate_kwarg   )Starcoder2Configc                       s@   e Zd Zdef fddZdeeej  dejfddZ	  Z
S )Starcoder2MLPconfigc                    sT   t    |j}tj||j|jd| _tj|j||jd| _t	|j
 | _|j| _d S )Nbias)super__init__hidden_sizer   Linearintermediate_sizeuse_biasc_fcc_projr   
hidden_actactresidual_dropout)selfr    	embed_dim	__class__ f/home/ubuntu/.local/lib/python3.10/site-packages/transformers/models/starcoder2/modeling_starcoder2.pyr$   6   s   
zStarcoder2MLP.__init__hidden_statesreturnc                 C   s8   |  |}| |}| |}tjj|| j| jd}|S )Nptraining)r)   r,   r*   r   
functionaldropoutr-   r8   )r.   r4   r2   r2   r3   forward>   s
   


zStarcoder2MLP.forward)__name__
__module____qualname__r   r$   r   tupletorchFloatTensorr;   __classcell__r2   r2   r0   r3   r   5   s    &r   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..N   dim)shaper@   cat)xx1x2r2   r2   r3   rotate_halfF   s   rL   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.
    )	unsqueezerL   )qkcossinposition_idsunsqueeze_dimq_embedk_embedr2   r2   r3   apply_rotary_pos_embM   s
   

rV   r4   n_repr5   c                 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)rG   expandreshape)r4   rW   batchnum_key_value_headsslenhead_dimr2   r2   r3   	repeat_kvh   s
   0r^           modulequerykeyvalueattention_maskscalingr:   kwargsc                 K   s   t || j}t || j}	t||dd| }
|d ur3|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 )NrD   r   rC   )rF   dtyper6   r   )r^   num_key_value_groupsr@   matmul	transposerG   r   r9   softmaxfloat32torh   r:   r8   
contiguous)r`   ra   rb   rc   rd   re   r:   rf   
key_statesvalue_statesattn_weightscausal_maskattn_outputr2   r2   r3   eager_attention_forwardt   s   
&ru   c                       s   e Zd ZdZddedee f fddZeddd	d
		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dZ  ZS )Starcoder2Attentionz=Multi-headed attention from 'Attention Is All You Need' paperNr    	layer_idxc                    s   t    || _|| _t|dd p|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| _|j| _d S )Nr]   g      Tr!   )r#   r$   r    rw   getattrr%   num_attention_headsr]   r[   ri   re   attention_dropout	is_causalr   r&   r(   q_projk_projv_projo_projr-   r.   r    rw   r0   r2   r3   r$      s   
zStarcoder2Attention.__init__past_key_valuepast_key_values4.58new_nameversionr4   position_embeddingsrd   cache_positionrf   r5   c                 K   sF  |j d d }g |d| jR }| ||dd}	| ||dd}
| ||dd}|\}}t|	|
||\}	}
|d urW|||d}||
|| j	|\}
}t
}| jjdkret| jj }|| |	|
||f| jsqdn| j| jt| jdd d|\}}|jg |dR   }| |}tjj|| j| jd	}||fS )
NrC   r   rD   )rQ   rP   r   eagerr_   sliding_window)r:   re   r   r6   )rG   r]   r|   viewrk   r}   r~   rV   updaterw   ru   r    _attn_implementationr   r8   rz   re   rx   rY   ro   r   r   r9   r:   r-   )r.   r4   r   rd   r   r   rf   input_shapehidden_shapequery_statesrp   rq   rP   rQ   cache_kwargsattention_interfacert   rr   r2   r2   r3   r;      s@   
	


zStarcoder2Attention.forwardN)NN)r<   r=   r>   __doc__r   r   intr$   r   r@   Tensorr?   r	   
LongTensorr   r   r;   rB   r2   r2   r0   r3   rv      s*    rv   c                       s   e Zd Zdedef fddZedddd							
				ddejde	ej de	ej
 de	e de	e de	ej
 de	eejejf  dee dejfddZ  ZS )Starcoder2DecoderLayerr    rw   c                    sV   t    |j| _t||d| _t|| _tj|j|j	d| _
tj|j|j	d| _d S )N)r    rw   eps)r#   r$   r%   rv   	self_attnr   mlpr   	LayerNormnorm_epsiloninput_layernormpost_attention_layernormr   r0   r2   r3   r$      s   

zStarcoder2DecoderLayer.__init__r   r   r   r   NFr4   rd   rR   	use_cacher   r   rf   r5   c              
   K   s^   |}	|  |}| jd|||||||d|\}}
|	| }|}	| |}| |}|	| }|S )N)r4   rd   rR   r   r   r   r   r2   )r   r   r   r   )r.   r4   rd   rR   r   r   r   r   rf   residual_r2   r2   r3   r;      s&   




zStarcoder2DecoderLayer.forward)NNNFNN)r<   r=   r>   r   r   r$   r   r@   r   r   r   r	   boolr?   r   r   r;   rB   r2   r2   r0   r3   r      s8    	
r   c                       sD   e Zd ZU ejed< ddef fddZe e	dd Z
  ZS )	Starcoder2RotaryEmbeddinginv_freqNr    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defaultr   F)
persistent)r#   r$   hasattr
isinstancer   dictgetr   max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenr    r   rope_init_fnattention_scalingregister_bufferr   original_inv_freq)r.   r    devicer   r0   r2   r3   r$     s   
z"Starcoder2RotaryEmbedding.__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r6|jjdkr6|jjnd}t	j
|dd+ | |  dd}t	j||fdd	}| | j }| | j }	W d    n1 smw   Y  |j|jd
|	j|jd
fS )Nr   rC   r   mpscpuF)device_typeenabledrD   rE   )rh   )r   floatrX   rG   rn   r   r   r   strr@   autocastrk   rH   rP   r   rQ   rh   )
r.   rI   rR   inv_freq_expandedposition_ids_expandedr   freqsembrP   rQ   r2   r2   r3   r;     s   0&z!Starcoder2RotaryEmbedding.forwardr   )r<   r=   r>   r@   r   __annotations__r   r$   no_gradr   r;   rB   r2   r2   r0   r3   r      s   
 
r   c                   @   sH   e Zd ZU eed< dZdZdgZdgZdZ	dZ
dZdZdZeedZdS )Starcoder2PreTrainedModelr    modelTr   r   )r4   
attentionsN)r<   r=   r>   r   r   base_model_prefixsupports_gradient_checkpointing_no_split_modules_skip_keys_device_placement_supports_flash_attn_supports_sdpa_supports_flex_attn_can_compile_fullgraph_supports_attention_backendr   rv   _can_record_outputsr2   r2   r2   r3   r   "  s   
 
r   c                       s   e Zd Zdef fddZe							ddeej deej	 deej dee
eeej f  d	eej d
ee deej dee defddZ  ZS )Starcoder2Modelr    c                    s   t     j| _ j| _t j j| j| _t	 fddt
 jD | _tj j jd| _t d| _d| _ j| _|   d S )Nc                    s   g | ]}t  |qS r2   )r   ).0rw   r    r2   r3   
<listcomp>>  s    z,Starcoder2Model.__init__.<locals>.<listcomp>r   r   F)r#   r$   pad_token_idpadding_idx
vocab_sizer   	Embeddingr%   embed_tokens
ModuleListrangenum_hidden_layerslayersr   r   normr   
rotary_embgradient_checkpointingembedding_dropout	post_initr.   r    r0   r   r3   r$   7  s   zStarcoder2Model.__init__N	input_idsrd   rR   r   inputs_embedsr   r   rf   r5   c              
   K   s4  |d u |d uA rt d|d u r| |}|r!|d u r!t| jd}|d u r=|d ur-| nd}	tj|	|	|jd  |jd}|d u rF|	d}| jj
d u rNtnt}
|
| j|||||d}|}tjj|| j| jd}| ||}| jd | jj D ]}||f||||||d|}qw| |}t||r|d	S d d	S )
Nz:You must specify exactly one of input_ids or inputs_embedsr   r   r   )r   )r    input_embedsrd   r   r   rR   r6   )rd   rR   r   r   r   r   )last_hidden_stater   )
ValueErrorr   r
   r    get_seq_lengthr@   arangerG   r   rM   r   r   r   r   r9   r:   r   r8   r   r   r   r   r   )r.   r   rd   rR   r   r   r   r   rf   past_seen_tokensmask_functionrs   r4   r   decoder_layerr2   r2   r3   r;   H  s^   

	

zStarcoder2Model.forward)NNNNNNN)r<   r=   r>   r   r$   r   r   r@   r   r   r   r	   listrA   r   r   r   r   r;   rB   r2   r2   r0   r3   r   5  s:    	
r   c                       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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dZ  ZS )Starcoder2ForCausalLMzlm_head.weightlm_headcolwise_repr4   logitsc                    s@   t  | t|| _|j| _tj|j|jdd| _| 	  d S )NFr!   )
r#   r$   r   r   r   r   r&   r%   r   r   r   r0   r2   r3   r$     s
   
zStarcoder2ForCausalLM.__init__Nr   r   rd   rR   r   r   labelsr   r   logits_to_keeprf   r5   c
              
   K   s   | j d|||||||d|
}|j}t|	trt|	 dn|	}| |dd|ddf }d}|durB| jd||| jjd|
}t	|||j
|j|jdS )a  
        Example:

        ```python
        >>> from transformers import AutoTokenizer, Starcoder2ForCausalLM

        >>> model = Starcoder2ForCausalLM.from_pretrained("meta-starcoder2/Starcoder2-2-7b-hf")
        >>> tokenizer = AutoTokenizer.from_pretrained("meta-starcoder2/Starcoder2-2-7b-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   rd   rR   r   r   r   r   N)r   r   r   )lossr   r   r4   r   r2   )r   r   r   r   slicer   loss_functionr    r   r   r   r4   r   )r.   r   rd   rR   r   r   r   r   r   r   rf   outputsr4   slice_indicesr   r   r2   r2   r3   r;     s0    zStarcoder2ForCausalLM.forward)	NNNNNNNNr   )r<   r=   r>   _tied_weights_keys_tp_plan_pp_planr$   r   r   r   r@   r   r   r	   rA   r   r   r   r   r   r   r;   rB   r2   r2   r0   r3   r     sN    		
r   c                   @      e Zd ZdS )#Starcoder2ForSequenceClassificationNr<   r=   r>   r2   r2   r2   r3   r         r   c                   @   r   ) Starcoder2ForTokenClassificationNr   r2   r2   r2   r3   r    r  r  )r   r   r   r   r  )Nr   )r_   )=typingr   r   r   r@   r   transformers.utils.genericr   activationsr   cache_utilsr	   r
   
generationr   masking_utilsr   r   modeling_flash_attention_utilsr   modeling_layersr   r   r   modeling_outputsr   r   modeling_rope_utilsr   r   modeling_utilsr   r   processing_utilsr   utilsr   r   r   utils.deprecationr   configuration_starcoder2r   Moduler   rL   rV   r   r   r^   r   ru   rv   r   r   r   r   r   r   r  __all__r2   r2   r2   r3   <module>   sf   

D,$UK