
    rh'                        d Z ddlZddlZddlmZmZmZ ddlZddl	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 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 m!Z! ddl"m#Z#  e jH                  e%      Z& G d de
jN                        Z( G d de
jN                        Z)	 d7de
jN                  dejT                  dejT                  dejT                  deejT                     de+de+fdZ, G d de
jN                        Z- G d de
jN                        Z. G d d e
jN                        Z/ G d! d"e
jN                        Z0 G d# d$e
jN                        Z1 G d% d&e      Z2 G d' d(e
jN                        Z3e G d) d*e             Z4e G d+ d,e4             Z5 G d- d.e
jN                        Z6 ed/0       G d1 d2e4             Z7 ed30       G d4 d5e4             Z8g d6Z9y)8zPyTorch ViT model.    N)CallableOptionalUnion)nn)BCEWithLogitsLossCrossEntropyLossMSELoss   )ACT2FN)GradientCheckpointingLayer)BaseModelOutputBaseModelOutputWithPoolingImageClassifierOutputMaskedImageModelingOutput)ALL_ATTENTION_FUNCTIONSPreTrainedModel) find_pruneable_heads_and_indicesprune_linear_layer)auto_docstringlogging	torch_int   )	ViTConfigc            	            e Zd ZdZddededdf fdZdej                  de	d	e	dej                  fd
Z
	 	 ddej                  deej                     dedej                  fdZ xZS )ViTEmbeddingszb
    Construct the CLS token, position and patch embeddings. Optionally, also the mask token.
    configuse_mask_tokenreturnNc                 J   t         |           t        j                  t	        j
                  dd|j                              | _        |r4t        j                  t	        j                  dd|j                              nd | _	        t        |      | _        | j                  j                  }t        j                  t	        j
                  d|dz   |j                              | _        t        j                  |j                        | _        |j"                  | _        || _        y )Nr   )super__init__r   	Parametertorchrandnhidden_size	cls_tokenzeros
mask_tokenViTPatchEmbeddingspatch_embeddingsnum_patchesposition_embeddingsDropouthidden_dropout_probdropout
patch_sizer   )selfr   r   r+   	__class__s       w/var/www/html/ai-insurance-compliance-backend/venv/lib/python3.12/site-packages/transformers/models/vit/modeling_vit.pyr!   zViTEmbeddings.__init__0   s    ekk!Q8J8J&KLQ_",,u{{1a9K9K'LMei 26 :++77#%<<A{QPVPbPb0c#d zz&"<"<= ++    
embeddingsheightwidthc                    |j                   d   dz
  }| j                  j                   d   dz
  }t        j                  j	                         s||k(  r||k(  r| j                  S | j                  ddddf   }| j                  ddddf   }|j                   d   }|| j
                  z  }	|| j
                  z  }
t        |dz        }|j                  d|||      }|j                  dddd      }t        j                  j                  ||	|
fdd	
      }|j                  dddd      j                  dd|      }t        j                  ||fd      S )a   
        This method allows to interpolate the pre-trained position encodings, to be able to use the model on higher resolution
        images. This method is also adapted to support torch.jit tracing.

        Adapted from:
        - https://github.com/facebookresearch/dino/blob/de9ee3df6cf39fac952ab558447af1fa1365362a/vision_transformer.py#L174-L194, and
        - https://github.com/facebookresearch/dinov2/blob/e1277af2ba9496fbadf7aec6eba56e8d882d1e35/dinov2/models/vision_transformer.py#L179-L211
        r   N      ?r   r
      bicubicF)sizemodealign_cornersdim)shaper,   r#   jit
is_tracingr0   r   reshapepermuter   
functionalinterpolateviewcat)r1   r5   r6   r7   r+   num_positionsclass_pos_embedpatch_pos_embedrA   
new_height	new_widthsqrt_num_positionss               r3   interpolate_pos_encodingz&ViTEmbeddings.interpolate_pos_encoding<   s`    !&&q)A-0066q9A= yy##%+*F6UZ?+++221bqb59221ab59r"t.
T__,	&}c'9:)11!5GI[]`a)11!Q1=--33i(	 4 
 *11!Q1=BB1b#Nyy/?;CCr4   pixel_valuesbool_masked_posrQ   c                    |j                   \  }}}}| j                  ||      }|Z|j                   d   }	| j                  j                  ||	d      }
|j	                  d      j                  |
      }|d|z
  z  |
|z  z   }| j                  j                  |dd      }t        j                  ||fd      }|r|| j                  |||      z   }n|| j                  z   }| j                  |      }|S )N)rQ   r   r9         ?r@   )rB   r*   r(   expand	unsqueezetype_asr&   r#   rJ   rQ   r,   r/   )r1   rR   rS   rQ   
batch_sizenum_channelsr6   r7   r5   
seq_lengthmask_tokensmask
cls_tokenss                r3   forwardzViTEmbeddings.forwardd   s    3?2D2D/
L&%**<Rj*k
&#))!,J//00ZLK",,R088ED#sTz2[45GGJ ^^**:r2>
YY
J7Q?
 $#d&C&CJPVX]&^^J#d&>&>>J\\*-
r4   FNF)__name__
__module____qualname____doc__r   boolr!   r#   TensorintrQ   r   
BoolTensorr_   __classcell__r2   s   @r3   r   r   +   s    
y 
$ 
4 
&D5<< &D &DUX &D]b]i]i &DV 7;).	ll "%"2"23 #'	
 
r4   r   c                   `     e Zd ZdZ fdZddej                  dedej                  fdZ xZ	S )r)   z
    This class turns `pixel_values` of shape `(batch_size, num_channels, height, width)` into the initial
    `hidden_states` (patch embeddings) of shape `(batch_size, seq_length, hidden_size)` to be consumed by a
    Transformer.
    c                    t         |           |j                  |j                  }}|j                  |j
                  }}t        |t        j                  j                        r|n||f}t        |t        j                  j                        r|n||f}|d   |d   z  |d   |d   z  z  }|| _        || _        || _        || _
        t        j                  ||||      | _        y )Nr   r   )kernel_sizestride)r    r!   
image_sizer0   rZ   r%   
isinstancecollectionsabcIterabler+   r   Conv2d
projection)r1   r   rp   r0   rZ   r%   r+   r2   s          r3   r!   zViTPatchEmbeddings.__init__   s    !'!2!2F4E4EJ
$*$7$79K9Kk#-j+//:R:R#SZZdfpYq
#-j+//:R:R#SZZdfpYq
!!}
15*Q-:VW=:XY$$(&))L+:^hir4   rR   rQ   r   c                    |j                   \  }}}}|| j                  k7  rt        d| j                   d| d      |sV|| j                  d   k7  s|| j                  d   k7  r2t        d| d| d| j                  d    d| j                  d    d		      | j	                  |      j                  d
      j                  dd
      }|S )NzoMake sure that the channel dimension of the pixel values match with the one set in the configuration. Expected z	 but got .r   r   zInput image size (*z) doesn't match model (z).r;   )rB   rZ   
ValueErrorrp   rv   flatten	transpose)r1   rR   rQ   rY   rZ   r6   r7   r5   s           r3   r_   zViTPatchEmbeddings.forward   s    2>2D2D/
L&%4,,,!../yaI  (++u8J/J (% 9+,Adooa.@-AE  __\2::1=GG1M
r4   r`   )
rb   rc   rd   re   r!   r#   rg   rf   r_   rj   rk   s   @r3   r)   r)      s3    jELL D ]b]i]i r4   r)   modulequerykeyvalueattention_maskscalingr/   c                    t        j                  ||j                  dd            |z  }t        j                  j                  |dt         j                        j                  |j                        }t        j                  j                  ||| j                        }|||z  }t        j                  ||      }	|	j                  dd      j                         }	|	|fS )Nr9   )rA   dtype)ptrainingr   r;   )r#   matmulr|   r   rG   softmaxfloat32tor   r/   r   
contiguous)
r}   r~   r   r   r   r   r/   kwargsattn_weightsattn_outputs
             r3   eager_attention_forwardr      s     <<s}}R'<=GL ==((2U]](SVVW\WbWbcL ==((6??([L !#n4,,|U3K''1-88:K$$r4   c            
            e Zd Zdeddf fdZ	 	 ddeej                     dede	e
ej                  ej                  f   e
ej                     f   fdZ xZS )	ViTSelfAttentionr   r   Nc                 2   t         |           |j                  |j                  z  dk7  r2t	        |d      s&t        d|j                   d|j                   d      || _        |j                  | _        t        |j                  |j                  z        | _        | j                  | j                  z  | _	        |j                  | _        | j                  dz  | _        d| _        t        j                  |j                  | j                  |j                         | _        t        j                  |j                  | j                  |j                         | _        t        j                  |j                  | j                  |j                         | _        y )	Nr   embedding_sizezThe hidden size z4 is not a multiple of the number of attention heads rx   g      F)bias)r    r!   r%   num_attention_headshasattrrz   r   rh   attention_head_sizeall_head_sizeattention_probs_dropout_probdropout_probr   	is_causalr   Linearqkv_biasr~   r   r   r1   r   r2   s     r3   r!   zViTSelfAttention.__init__   sF    : ::a?PVXhHi"6#5#5"6 7334A7 
 #)#=#= #&v'9'9F<V<V'V#W !558P8PP"??//5YYv1143E3EFOO\
99V//1C1C&//ZYYv1143E3EFOO\
r4   	head_maskoutput_attentionsc           
         |j                   \  }}}| j                  |      j                  |d| j                  | j                        j                  dd      }| j                  |      j                  |d| j                  | j                        j                  dd      }| j                  |      j                  |d| j                  | j                        j                  dd      }	t        }
| j                  j                  dk7  rN| j                  j                  dk(  r|rt        j                  d       nt        | j                  j                     }
 |
| |	|||| j                  | j                  | j                   sdn| j"                        \  }}|j%                         d d	 | j&                  fz   }|j)                  |      }|r||f}|S |f}|S )
Nr9   r   r;   eagersdpaz`torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to eager attention. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.        )r   r   r/   r   )rB   r   rI   r   r   r|   r   r~   r   r   _attn_implementationloggerwarning_oncer   r   r   r   r   r=   r   rE   )r1   hidden_statesr   r   rY   r[   _	key_layervalue_layerquery_layerattention_interfacecontext_layerattention_probsnew_context_layer_shapeoutputss                  r3   r_   zViTSelfAttention.forward   s    %2$7$7!
JHH]#T*b$":":D<T<TUYq!_ 	 JJ}%T*b$":":D<T<TUYq!_ 	 JJ}%T*b$":":D<T<TUYq!_ 	 )@;;++w6{{//69>O##L
 '>dkk>^>^&_#)<nnLL#}}C$2C2C	*
& #0"4"4"6s";t?Q?Q>S"S%--.EF6G=/2 O\M]r4   ra   )rb   rc   rd   r   r!   r   r#   rg   rf   r   tupler_   rj   rk   s   @r3   r   r      sr    ]y ]T ]. -1"'	1 ELL)1  	1
 
uU\\5<</0%2EE	F1r4   r   c                   |     e Zd ZdZdeddf fdZdej                  dej                  dej                  fdZ xZ	S )	ViTSelfOutputz
    The residual connection is defined in ViTLayer instead of here (as is the case with other models), due to the
    layernorm applied before each block.
    r   r   Nc                     t         |           t        j                  |j                  |j                        | _        t        j                  |j                        | _        y N)	r    r!   r   r   r%   denser-   r.   r/   r   s     r3   r!   zViTSelfOutput.__init__  sB    YYv1163E3EF
zz&"<"<=r4   r   input_tensorc                 J    | j                  |      }| j                  |      }|S r   r   r/   r1   r   r   s      r3   r_   zViTSelfOutput.forward  s$    

=1]3r4   )
rb   rc   rd   re   r   r!   r#   rg   r_   rj   rk   s   @r3   r   r     sD    
>y >T >
U\\  RWR^R^ r4   r   c                        e Zd Zdeddf fdZdee   ddfdZ	 	 ddej                  de
ej                     d	edeeej                  ej                  f   eej                     f   fd
Z xZS )ViTAttentionr   r   Nc                     t         |           t        |      | _        t	        |      | _        t               | _        y r   )r    r!   r   	attentionr   outputsetpruned_headsr   s     r3   r!   zViTAttention.__init__$  s0    )&1#F+Er4   headsc                 >   t        |      dk(  ry t        || j                  j                  | j                  j                  | j
                        \  }}t        | j                  j                  |      | j                  _        t        | j                  j                  |      | j                  _        t        | j                  j                  |      | j                  _	        t        | j                  j                  |d      | j                  _        | j                  j                  t        |      z
  | j                  _        | j                  j                  | j                  j                  z  | j                  _        | j
                  j                  |      | _        y )Nr   r   r@   )lenr   r   r   r   r   r   r~   r   r   r   r   r   union)r1   r   indexs      r3   prune_headszViTAttention.prune_heads*  s   u:?74>>55t~~7Y7Y[_[l[l
u
  2$..2F2FN/0B0BEJ1$..2F2FN.t{{/@/@%QO .2^^-O-ORUV[R\-\*'+~~'I'IDNNLnLn'n$ --33E:r4   r   r   r   c                 h    | j                  |||      }| j                  |d   |      }|f|dd  z   }|S )Nr   r   )r   r   )r1   r   r   r   self_outputsattention_outputr   s          r3   r_   zViTAttention.forward<  sE     ~~mY@QR;;|AF#%QR(88r4   ra   )rb   rc   rd   r   r!   r   rh   r   r#   rg   r   rf   r   r   r_   rj   rk   s   @r3   r   r   #  s    "y "T ";S ;d ;* -1"'	|| ELL)  	
 
uU\\5<</0%2EE	Fr4   r   c                   `     e Zd Zdeddf fdZdej                  dej                  fdZ xZS )ViTIntermediater   r   Nc                    t         |           t        j                  |j                  |j
                        | _        t        |j                  t              rt        |j                     | _        y |j                  | _        y r   )r    r!   r   r   r%   intermediate_sizer   rq   
hidden_actstrr   intermediate_act_fnr   s     r3   r!   zViTIntermediate.__init__K  s]    YYv1163K3KL
f''-'-f.?.?'@D$'-'8'8D$r4   r   c                 J    | j                  |      }| j                  |      }|S r   )r   r   )r1   r   s     r3   r_   zViTIntermediate.forwardS  s&    

=100?r4   	rb   rc   rd   r   r!   r#   rg   r_   rj   rk   s   @r3   r   r   J  s1    9y 9T 9U\\ ell r4   r   c                   x     e Zd Zdeddf fdZdej                  dej                  dej                  fdZ xZS )	ViTOutputr   r   Nc                     t         |           t        j                  |j                  |j
                        | _        t        j                  |j                        | _	        y r   )
r    r!   r   r   r   r%   r   r-   r.   r/   r   s     r3   r!   zViTOutput.__init__[  sB    YYv779K9KL
zz&"<"<=r4   r   r   c                 T    | j                  |      }| j                  |      }||z   }|S r   r   r   s      r3   r_   zViTOutput.forward`  s.    

=1]3%4r4   r   rk   s   @r3   r   r   Z  s?    >y >T >
U\\  RWR^R^ r4   r   c                        e Zd ZdZdeddf fdZ	 	 d
dej                  deej                     de	de
eej                  ej                  f   eej                     f   fd	Z xZS )ViTLayerz?This corresponds to the Block class in the timm implementation.r   r   Nc                 r   t         |           |j                  | _        d| _        t	        |      | _        t        |      | _        t        |      | _	        t        j                  |j                  |j                        | _        t        j                  |j                  |j                        | _        y )Nr   eps)r    r!   chunk_size_feed_forwardseq_len_dimr   r   r   intermediater   r   r   	LayerNormr%   layer_norm_epslayernorm_beforelayernorm_afterr   s     r3   r!   zViTLayer.__init__l  s    '-'E'E$%f-+F3' "V-?-?VEZEZ [!||F,>,>FDYDYZr4   r   r   r   c                     | j                  | j                  |      ||      }|d   }|dd  }||z   }| j                  |      }| j                  |      }| j	                  ||      }|f|z   }|S )N)r   r   r   )r   r   r   r   r   )r1   r   r   r   self_attention_outputsr   r   layer_outputs           r3   r_   zViTLayer.forwardv  s     "&!!-0/ "0 "

 2!4(, )=8 ++M:((6 {{<?/G+r4   ra   )rb   rc   rd   re   r   r!   r#   rg   r   rf   r   r   r_   rj   rk   s   @r3   r   r   i  s    I[y [T [ -1"'	|| ELL)  	
 
uU\\5<</0%2EE	Fr4   r   c                        e Zd Zdeddf fdZ	 	 	 	 ddej                  deej                     deded	ede	e
ef   fd
Z xZS )
ViTEncoderr   r   Nc                     t         |           || _        t        j                  t        |j                        D cg c]  }t        |       c}      | _        d| _	        y c c}w ra   )
r    r!   r   r   
ModuleListrangenum_hidden_layersr   layergradient_checkpointing)r1   r   r   r2   s      r3   r!   zViTEncoder.__init__  sN    ]]eFD\D\>]#^HV$4#^_
&+# $_s   A#r   r   r   output_hidden_statesreturn_dictc                    |rdnd }|rdnd }t        | j                        D ]1  \  }}	|r||fz   }|||   nd }
 |	||
|      }|d   }|s)||d   fz   }3 |r||fz   }|st        d |||fD              S t        |||      S )N r   r   c              3   &   K   | ]	  }||  y wr   r   ).0vs     r3   	<genexpr>z%ViTEncoder.forward.<locals>.<genexpr>  s     mq_`_lms   )last_hidden_stater   
attentions)	enumerater   r   r   )r1   r   r   r   r   r   all_hidden_statesall_self_attentionsilayer_modulelayer_head_masklayer_outputss               r3   r_   zViTEncoder.forward  s     #7BD$5b4(4 	POA|#$58H$H!.7.CilO(IZ[M)!,M &9]1=M<O&O#	P   1]4D Dm]4EGZ$[mmm++*
 	
r4   )NFFT)rb   rc   rd   r   r!   r#   rg   r   rf   r   r   r   r_   rj   rk   s   @r3   r   r     sz    ,y ,T , -1"'%* !
||!
 ELL)!
  	!

 #!
 !
 
uo%	&!
r4   r   c                       e Zd ZU eed<   dZdZdZddgZdZ	dZ
dZdZdeej                  ej                   ej"                  f   dd	fd
Zy	)ViTPreTrainedModelr   vitrR   Tr   r   r}   r   Nc                    t        |t        j                  t        j                  f      rt        j                  j                  |j                  j                  j                  t        j                        d| j                  j                        j                  |j                  j                        |j                  _        |j                  %|j                  j                  j                          yyt        |t        j                         rJ|j                  j                  j                          |j                  j                  j#                  d       yt        |t$              rdt        j                  j                  |j&                  j                  j                  t        j                        d| j                  j                        j                  |j&                  j                        |j&                  _        t        j                  j                  |j(                  j                  j                  t        j                        d| j                  j                        j                  |j(                  j                        |j(                  _        |j*                  %|j*                  j                  j                          yyy)zInitialize the weightsr   )meanstdNrU   )rq   r   r   ru   inittrunc_normal_weightdatar   r#   r   r   initializer_ranger   r   zero_r   fill_r   r,   r&   r(   )r1   r}   s     r3   _init_weightsz ViTPreTrainedModel._init_weights  s   fryy"))45 "$!6!6""%%emm43DKKDaDa "7 "b$$% MM {{&  &&( '-KK""$MM$$S)..0gg.C.C**//225==AKK11 /D / b++112	 &&+ %'GG$9$9  %%((7KK11 %: % b!!''(	 !   ,!!&&,,. - /r4   )rb   rc   rd   r   __annotations__base_model_prefixmain_input_namesupports_gradient_checkpointing_no_split_modules_supports_sdpa_supports_flash_attn_supports_flex_attn_supports_attention_backendr   r   r   ru   r   r  r   r4   r3   r  r    sg    $O&*#(*5N"&/E"))RYY*L$M /RV /r4   r  c                       e Zd Zddededef fdZdefdZdee	e
e	   f   ddfd	Ze	 	 	 	 	 	 	 dd
eej                     deej                      deej                     dee   dee   dee   dee   deeef   fd       Z xZS )ViTModelr   add_pooling_layerr   c                    t         |   |       || _        t        ||      | _        t        |      | _        t        j                  |j                  |j                        | _        |rt        |      nd| _        | j                          y)z
        add_pooling_layer (bool, *optional*, defaults to `True`):
            Whether to add a pooling layer
        use_mask_token (`bool`, *optional*, defaults to `False`):
            Whether to use a mask token for masked image modeling.
        )r   r   N)r    r!   r   r   r5   r   encoderr   r   r%   r   	layernorm	ViTPoolerpooler	post_init)r1   r   r  r   r2   s       r3   r!   zViTModel.__init__  sm     	 '~N!&)f&8&8f>S>ST+<i'$ 	r4   r   c                 .    | j                   j                  S r   )r5   r*   )r1   s    r3   get_input_embeddingszViTModel.get_input_embeddings  s    ///r4   heads_to_pruneNc                     |j                         D ]7  \  }}| j                  j                  |   j                  j	                  |       9 y)z
        Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer} See base
        class PreTrainedModel
        N)itemsr  r   r   r   )r1   r$  r   r   s       r3   _prune_headszViTModel._prune_heads   sE    
 +002 	CLE5LLu%//;;EB	Cr4   rR   rS   r   r   r   rQ   r   c                    ||n| j                   j                  }||n| j                   j                  }||n| j                   j                  }|t	        d      | j                  || j                   j                        }| j                  j                  j                  j                  j                  }|j                  |k7  r|j                  |      }| j                  |||      }	| j                  |	||||      }
|
d   }| j                  |      }| j                  | j                  |      nd}|s|||fn|f}||
dd z   S t!        |||
j"                  |
j$                        S )z
        bool_masked_pos (`torch.BoolTensor` of shape `(batch_size, num_patches)`, *optional*):
            Boolean masked positions. Indicates which patches are masked (1) and which aren't (0).
        Nz You have to specify pixel_values)rS   rQ   )r   r   r   r   r   r   )r   pooler_outputr   r   )r   r   r   use_return_dictrz   get_head_maskr   r5   r*   rv   r
  r   r   r  r  r   r   r   r   )r1   rR   rS   r   r   r   rQ   r   expected_dtypeembedding_outputencoder_outputssequence_outputpooled_outputhead_outputss                 r3   r_   zViTModel.forward  s    2C1N-TXT_T_TqTq$8$D $++JjJj 	 &1%<k$++B]B]?@@ &&y$++2O2OP	 99DDKKQQ/'??>:L??/Tl + 
 ,,/!5# ' 
 *!,..98<8OO4UY?L?XO];_n^pL/!""555)-')77&11	
 	
r4   )TFNNNNNNN)rb   rc   rd   r   rf   r!   r)   r#  dictrh   listr'  r   r   r#   rg   ri   r   r   r   r_   rj   rk   s   @r3   r  r    s    y T Z^ &0&8 0C4T#Y+? CD C  046:,0,0/337&*;
u||,;
 "%"2"23;
 ELL)	;

 $D>;
 'tn;
 #+4.;
 d^;
 
u00	1;
 ;
r4   r  c                   *     e Zd Zdef fdZd Z xZS )r  r   c                     t         |           t        j                  |j                  |j
                        | _        t        |j                     | _	        y r   )
r    r!   r   r   r%   pooler_output_sizer   r   
pooler_act
activationr   s     r3   r!   zViTPooler.__init__H  s>    YYv1163L3LM
 !2!23r4   c                 \    |d d df   }| j                  |      }| j                  |      }|S )Nr   )r   r9  )r1   r   first_token_tensorr0  s       r3   r_   zViTPooler.forwardM  s6     +1a40

#566r4   )rb   rc   rd   r   r!   r_   rj   rk   s   @r3   r  r  G  s    4y 4
r4   r  ac  
    ViT Model with a decoder on top for masked image modeling, as proposed in [SimMIM](https://huggingface.co/papers/2111.09886).

    <Tip>

    Note that we provide a script to pre-train this model on custom data in our [examples
    directory](https://github.com/huggingface/transformers/tree/main/examples/pytorch/image-pretraining).

    </Tip>
    )custom_introc                        e Zd Zdeddf fdZe	 	 	 	 	 	 	 ddeej                     deej                     deej                     dee
   d	ee
   d
ee
   dee
   deeef   fd       Z xZS )ViTForMaskedImageModelingr   r   Nc                 N   t         |   |       t        |dd      | _        t	        j
                  t	        j                  |j                  |j                  dz  |j                  z  d      t	        j                  |j                              | _        | j                          y )NFT)r  r   r;   r   )in_channelsout_channelsrn   )r    r!   r  r  r   
Sequentialru   r%   encoder_striderZ   PixelShuffledecoderr!  r   s     r3   r!   z"ViTForMaskedImageModeling.__init__c  s     FeDQ}}II"..#22A58K8KK
 OOF112
 	r4   rR   rS   r   r   r   rQ   r   c           	         ||n| j                   j                  }|g| j                   j                  | j                   j                  k7  r:t	        d| j                   j                   d| j                   j                   d      | j                  |||||||      }|d   }	|	ddddf   }	|	j                  \  }
}}t        j                  |dz        x}}|	j                  dd	d      j                  |
|||      }	| j                  |	      }d}|| j                   j                  | j                   j                  z  }|j                  d
||      }|j                  | j                   j                  d      j                  | j                   j                  d	      j                  d      j                         }t         j"                  j%                  ||d      }||z  j'                         |j'                         dz   z  | j                   j(                  z  }|s|f|dd z   }||f|z   S |S t+        |||j,                  |j.                        S )a+  
        bool_masked_pos (`torch.BoolTensor` of shape `(batch_size, num_patches)`):
            Boolean masked positions. Indicates which patches are masked (1) and which aren't (0).

        Examples:
        ```python
        >>> from transformers import AutoImageProcessor, ViTForMaskedImageModeling
        >>> import torch
        >>> from PIL import Image
        >>> import requests

        >>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
        >>> image = Image.open(requests.get(url, stream=True).raw)

        >>> image_processor = AutoImageProcessor.from_pretrained("google/vit-base-patch16-224-in21k")
        >>> model = ViTForMaskedImageModeling.from_pretrained("google/vit-base-patch16-224-in21k")

        >>> num_patches = (model.config.image_size // model.config.patch_size) ** 2
        >>> pixel_values = image_processor(images=image, return_tensors="pt").pixel_values
        >>> # create random boolean mask of shape (batch_size, num_patches)
        >>> bool_masked_pos = torch.randint(low=0, high=2, size=(1, num_patches)).bool()

        >>> outputs = model(pixel_values, bool_masked_pos=bool_masked_pos)
        >>> loss, reconstructed_pixel_values = outputs.loss, outputs.reconstruction
        >>> list(reconstructed_pixel_values.shape)
        [1, 3, 224, 224]
        ```NzWhen `bool_masked_pos` is provided, `patch_size` must be equal to `encoder_stride` to ensure that the reconstructed image has the same dimensions as the input. Got `patch_size` = z and `encoder_stride` = rx   )rS   r   r   r   rQ   r   r   r   r:   r;   r9   none)	reductiongh㈵>)lossreconstructionr   r   )r   r*  r0   rC  rz   r  rB   mathfloorrF   rE   rE  rp   repeat_interleaverW   r   r   rG   l1_losssumrZ   r   r   r   )r1   rR   rS   r   r   r   rQ   r   r   r/  rY   sequence_lengthrZ   r6   r7   reconstructed_pixel_valuesmasked_im_lossr=   r]   reconstruction_lossr   s                        r3   r_   z!ViTForMaskedImageModeling.forwardt  sR   L &1%<k$++B]B]&DKK,B,BdkkF`F`,`&&*kk&<&<%==UVZVaVaVpVpUqqrt  ((+/!5%=#  
 "!* *!QR%04C4I4I1
O\OS$899)11!Q:BB:|]cejk &*\\/%B"&;;))T[[-C-CCD-55b$EO11$++2H2H!L""4;;#9#91=1	  #%--"7"7F`lr"7"s1D8==?488:PTCTUX\XcXcXpXppN02WQR[@F3A3M^%.YSYY(5!//))	
 	
r4   r2  )rb   rc   rd   r   r!   r   r   r#   rg   ri   rf   r   r   r   r_   rj   rk   s   @r3   r>  r>  V  s    y T "  046:,0,0/337&*Y
u||,Y
 "%"2"23Y
 ELL)	Y

 $D>Y
 'tnY
 #+4.Y
 d^Y
 
u//	0Y
 Y
r4   r>  a  
    ViT Model transformer with an image classification head on top (a linear layer on top of the final hidden state of
    the [CLS] token) e.g. for ImageNet.

    <Tip>

        Note that it's possible to fine-tune ViT on higher resolution images than the ones it has been trained on, by
        setting `interpolate_pos_encoding` to `True` in the forward of the model. This will interpolate the pre-trained
        position embeddings to the higher resolution.

    </Tip>
    c                        e Zd Zdeddf fdZe	 	 	 	 	 	 	 ddeej                     deej                     deej                     dee	   d	ee	   d
ee	   dee	   de
eef   fd       Z xZS )ViTForImageClassificationr   r   Nc                 .   t         |   |       |j                  | _        t        |d      | _        |j                  dkD  r*t        j                  |j                  |j                        nt        j                         | _	        | j                          y )NF)r  r   )r    r!   
num_labelsr  r  r   r   r%   Identity
classifierr!  r   s     r3   r!   z"ViTForImageClassification.__init__  ss      ++Fe< OUN_N_bcNc"))F$6$68I8IJikititiv 	r4   rR   r   labelsr   r   rQ   r   c                 b   ||n| j                   j                  }| j                  ||||||      }|d   }	| j                  |	dddddf         }
d}||j	                  |
j
                        }| j                   j                  | j                  dk(  rd| j                   _        nl| j                  dkD  rL|j                  t        j                  k(  s|j                  t        j                  k(  rd| j                   _        nd| j                   _        | j                   j                  dk(  rIt               }| j                  dk(  r& ||
j                         |j                               }n ||
|      }n| j                   j                  dk(  r=t               } ||
j                  d| j                        |j                  d            }n,| j                   j                  dk(  rt!               } ||
|      }|s|
f|dd z   }||f|z   S |S t#        ||
|j$                  |j&                  	      S )
a  
        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
            Labels for computing the image classification/regression loss. Indices should be in `[0, ...,
            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
        N)r   r   r   rQ   r   r   r   
regressionsingle_label_classificationmulti_label_classificationr9   )rI  logitsr   r   )r   r*  r  rY  r   deviceproblem_typerW  r   r#   longrh   r	   squeezer   rI   r   r   r   r   )r1   rR   r   rZ  r   r   rQ   r   r   r/  r_  rI  loss_fctr   s                 r3   r_   z!ViTForImageClassification.forward  s   " &1%<k$++B]B]((/!5%=#  
 "!*Aq!9:YYv}}-F{{''/??a'/;DKK,__q(fllejj.HFLL\a\e\eLe/LDKK,/KDKK,{{''<7"9??a'#FNN$4fnn6FGD#FF3D))-JJ+-B @&++b/R))-II,./Y,F)-)9TGf$EvE$!//))	
 	
r4   r2  )rb   rc   rd   r   r!   r   r   r#   rg   rf   r   r   r   r_   rj   rk   s   @r3   rU  rU    s    
y 
T 
  04,0)-,0/337&*A
u||,A
 ELL)A
 &	A

 $D>A
 'tnA
 #+4.A
 d^A
 
u++	,A
 A
r4   rU  )rU  r>  r  r  )r   ):re   collections.abcrr   rK  typingr   r   r   r#   torch.utils.checkpointr   torch.nnr   r   r	   activationsr   modeling_layersr   modeling_outputsr   r   r   r   modeling_utilsr   r   pytorch_utilsr   r   utilsr   r   r   configuration_vitr   
get_loggerrb   r   Moduler   r)   rg   floatr   r   r   r   r   r   r   r   r  r  r  r>  rU  __all__r   r4   r3   <module>rt     s      , ,    A A ! 9  G Q 7 7 ( 
		H	%UBII Up$ $\ %II%<<% 
% <<	%
 U\\*% % %<Fryy FRBII $$299 $Nbii  		 ') 'T(
 (
V &/ &/ &/R [
! [
 [
|		  	l
 2 l
l
^ O
 2 O
O
d gr4   