
    Uh|                     @   d Z ddlZddlmZmZmZmZ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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$Z$ ddl%m&Z&  e!jN                  e(      Z) G d dejT                        Z+ G d dejT                        Z,	 d>dejT                  dejZ                  dejZ                  dejZ                  deejZ                     de.de.fdZ/ G d dejT                        Z0 G d dejT                        Z1 G d d ejT                        Z2 G d! d"ejT                        Z3d?d#ejZ                  d$e.d%e4d&ejZ                  fd'Z5 G d( d)ejT                        Z6 G d* d+ejT                        Z7 G d, d-ejT                        Z8 G d. d/ejT                        Z9 G d0 d1ejT                        Z:e  G d2 d3e             Z;e  G d4 d5e;             Z< e d67       G d8 d9e;             Z= e d:7       G d; d<e;e$             Z>g d=Z?y)@zPyTorch DINOv2 model.    N)CallableDictListOptionalSetTupleUnion)nn)BCEWithLogitsLossCrossEntropyLossMSELoss   )ACT2FN)BackboneOutputBaseModelOutputBaseModelOutputWithPoolingImageClassifierOutput)ALL_ATTENTION_FUNCTIONSPreTrainedModel) find_pruneable_heads_and_indicesprune_linear_layer)auto_docstringlogging	torch_int)BackboneMixin   )Dinov2Configc                        e Zd ZdZ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j                  fdZ xZS )Dinov2EmbeddingszM
    Construct the CLS token, mask token, position and patch embeddings.
    configreturnNc                 z   t         |           t        j                  t	        j
                  dd|j                              | _        |j                  r8t        j                  t	        j                  d|j                              | _
        t        |      | _        | j                  j                  }t        j                  t	        j
                  d|dz   |j                              | _        t        j                  |j                         | _        |j$                  | _        |j                  | _        || _        y )Nr   )super__init__r
   	Parametertorchrandnhidden_size	cls_tokenuse_mask_tokenzeros
mask_tokenDinov2PatchEmbeddingspatch_embeddingsnum_patchesposition_embeddingsDropouthidden_dropout_probdropout
patch_sizer    )selfr    r/   	__class__s      |/var/www/catia.catastroantioquia-mas.com/valormas/lib/python3.12/site-packages/transformers/models/dinov2/modeling_dinov2.pyr$   zDinov2Embeddings.__init__*   s    ekk!Q8J8J&KL   ll5;;q&:L:L+MNDO 5f =++77#%<<A{QPVPbPb0c#d zz&"<"<= ++$33    
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      }|j                  }t        j                  j                  |j                  t        j                        |	|
fdd	
      j                  |      }|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 and interpolation at torch.float32 precision.

        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   Ng      ?r   r      bicubicF)sizemodealign_cornersdtypedim)shaper0   r&   jit
is_tracingr4   r   reshapepermuterD   r
   
functionalinterpolatetofloat32viewcat)r5   r9   r:   r;   r/   num_positionsclass_pos_embedpatch_pos_embedrF   
new_height	new_widthsqrt_num_positionstarget_dtypes                r7   interpolate_pos_encodingz)Dinov2Embeddings.interpolate_pos_encoding8   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u}}-i(	 4 

 "<"
  	 *11!Q1=BB1b#Nyy/?;CCr8   pixel_valuesbool_masked_posc                 D   |j                   \  }}}}| j                  j                  j                  j                  }| j                  |j                  |            }|d| j                  rXt        j                  |j                  d      | j                  j                  |j                        j                  d      |      }| j                  j                  |dd      }	t        j                  |	|fd      }|| j                  |||      z   }| j                  |      }|S )NrC   r=   r   r   rE   )rG   r.   
projectionweightrD   rN   r*   r&   where	unsqueezer,   r)   expandrQ   rY   r3   )
r5   rZ   r[   
batch_size_r:   r;   rX   r9   
cls_tokenss
             r7   forwardzDinov2Embeddings.forward`   s    '3'9'9$
Avu,,77>>DD**<???+NO
&4+>+>))"-t/A/A*BRBR/S/]/]^_/`blJ
 ^^**:r2>
YY
J7Q?
  $"?"?
FTY"ZZ
\\*-
r8   N)__name__
__module____qualname____doc__r   r$   r&   TensorintrY   r   re   __classcell__r6   s   @r7   r   r   %   s|    |  &D5<< &D &DUX &D]b]i]i &DPELL 8ELLCY ejeqeq r8   r   c                   Z     e Zd ZdZ fdZdej                  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$   
image_sizer4   num_channelsr(   
isinstancecollectionsabcIterabler/   r
   Conv2dr]   )r5   r    rs   r4   rt   r(   r/   r6   s          r7   r$   zDinov2PatchEmbeddings.__init__}   s    !'!2!2F4E4EJ
$*$7$79K9Kk#-j+//:R:R#SZZdfpYq
#-j+//:R:R#SZZdfpYq
!!}
15*Q-:VW=:XY$$(&))L+:^hir8   rZ   r!   c                     |j                   d   }|| j                  k7  rt        d| j                   d| d      | j                  |      j	                  d      j                  dd      }|S )Nr   zoMake sure that the channel dimension of the pixel values match with the one set in the configuration. Expected z	 but got .r>   )rG   rt   
ValueErrorr]   flatten	transpose)r5   rZ   rt   r9   s       r7   re   zDinov2PatchEmbeddings.forward   sz    #))!,4,,,!../yaI  __\2::1=GG1M
r8   )	rg   rh   ri   rj   r$   r&   rk   re   rm   rn   s   @r7   r-   r-   v   s)    jELL U\\ r8   r-   modulequerykeyvalueattention_maskscalingr3   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 )Nr=   )rF   rD   )ptrainingr   r>   )r&   matmulr~   r
   rL   softmaxrO   rN   rD   r3   r   
contiguous)
r   r   r   r   r   r   r3   kwargsattn_weightsattn_outputs
             r7   eager_attention_forwardr      s     <<s}}R'<=GL ==((2U]](SVVW\WbWbcL ==((6??([L !#n4,,|U3K''1-88:K$$r8   c            
            e Zd Zdeddf fdZdej                  dej                  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 )Dinov2SelfAttentionr    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 r{   g      Fbias)r#   r$   r(   num_attention_headshasattrr|   r    rl   attention_head_sizeall_head_sizeattention_probs_dropout_probdropout_probr   	is_causalr
   Linearqkv_biasr   r   r   r5   r    r6   s     r7   r$   zDinov2SelfAttention.__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\
r8   xc                     |j                         d d | j                  | j                  fz   }|j                  |      }|j	                  dddd      S )Nr=   r   r>   r   r   )r@   r   r   rP   rK   )r5   r   new_x_shapes      r7   transpose_for_scoresz(Dinov2SelfAttention.transpose_for_scores   sL    ffhsmt'?'?AYAY&ZZFF;yyAq!$$r8   	head_maskoutput_attentionsc           
         | j                  | j                  |            }| j                  | j                  |            }| j                  | j                  |            }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 )N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   r3   r   )r   r   r   r   r   r    _attn_implementationloggerwarning_oncer   r   r   r   r   r@   r   rJ   )r5   hidden_statesr   r   	key_layervalue_layerquery_layerattention_interfacecontext_layerattention_probsnew_context_layer_shapeoutputss               r7   re   zDinov2SelfAttention.forward   s=    --dhh}.EF	//

=0IJ//

=0IJ(?;;++w6{{//69>O##L
 '>dkk>^>^&_#)<nnLL#}}C$2C2C	*
& #0"4"4"6s";t?Q?Q>S"S%--.EF6G=/2 O\M]r8   NF)rg   rh   ri   r   r$   r&   rk   r   r   boolr	   r   re   rm   rn   s   @r7   r   r      s    ]| ] ](%ell %u|| % bg!(0(>!Z^!	uU\\5<</0%2EE	F!r8   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 )	Dinov2SelfOutputz
    The residual connection is defined in Dinov2Layer 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 rf   )	r#   r$   r
   r   r(   denser1   r2   r3   r   s     r7   r$   zDinov2SelfOutput.__init__   sB    YYv1163E3EF
zz&"<"<=r8   r   input_tensorc                 J    | j                  |      }| j                  |      }|S rf   )r   r3   )r5   r   r   s      r7   re   zDinov2SelfOutput.forward  s$    

=1]3r8   )
rg   rh   ri   rj   r   r$   r&   rk   re   rm   rn   s   @r7   r   r      sD    
>| > >
U\\  RWR^R^ r8   r   c                        e Zd Zdeddf fdZdee   ddfdZ	 	 ddej                  de
ej                     d	edeeej                  ej                  f   eej                     f   fd
Z xZS )Dinov2Attentionr    r!   Nc                     t         |           t        |      | _        t	        |      | _        t               | _        y rf   )r#   r$   r   	attentionr   outputsetpruned_headsr   s     r7   r$   zDinov2Attention.__init__
  s0    ,V4&v.Er8   headsc                 >   t        |      dk(  ry t        || j                  j                  | j                  j                  | j
                        \  }}t        | j                  j                  |      | j                  _        t        | j                  j                  |      | j                  _        t        | j                  j                  |      | j                  _	        t        | j                  j                  |d      | j                  _        | j                  j                  t        |      z
  | j                  _        | j                  j                  | j                  j                  z  | j                  _        | j
                  j                  |      | _        y )Nr   r   rE   )lenr   r   r   r   r   r   r   r   r   r   r   r   union)r5   r   indexs      r7   prune_headszDinov2Attention.prune_heads  s   u:?74>>55t~~7Y7Y[_[l[l
u
  2$..2F2FN/0B0BEJ1$..2F2FN.t{{/@/@%QO .2^^-O-ORUV[R\-\*'+~~'I'IDNNLnLn'n$ --33E:r8   r   r   r   c                 h    | j                  |||      }| j                  |d   |      }|f|dd  z   }|S )Nr   r   )r   r   )r5   r   r   r   self_outputsattention_outputr   s          r7   re   zDinov2Attention.forward"  sE     ~~mY@QR;;|AF#%QR(88r8   r   )rg   rh   ri   r   r$   r   rl   r   r&   rk   r   r   r	   r   re   rm   rn   s   @r7   r   r   	  s    "| " ";S ;d ;* -1"'	|| ELL)  	
 
uU\\5<</0%2EE	Fr8   r   c                   X     e Zd Zd fdZdej
                  dej
                  fdZ xZS )Dinov2LayerScaler!   c                     t         |           t        j                  |j                  t        j                  |j                        z        | _        y rf   )	r#   r$   r
   r%   layerscale_valuer&   onesr(   lambda1r   s     r7   r$   zDinov2LayerScale.__init__1  s8    ||F$;$;ejjI[I[>\$\]r8   hidden_statec                      || j                   z  S rf   )r   r5   r   s     r7   re   zDinov2LayerScale.forward5  s    dll**r8   r!   Nrg   rh   ri   r$   r&   rk   re   rm   rn   s   @r7   r   r   0  s$    ^+ELL +U\\ +r8   r   input	drop_probr   r!   c                    |dk(  s|s| S d|z
  }| j                   d   fd| j                  dz
  z  z   }|t        j                  || j                  | j
                        z   }|j                          | j                  |      |z  }|S )aF  
    Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).

    Comment by Ross Wightman: This is the same as the DropConnect impl I created for EfficientNet, etc networks,
    however, the original name is misleading as 'Drop Connect' is a different form of dropout in a separate paper...
    See discussion: https://github.com/tensorflow/tpu/issues/494#issuecomment-532968956 ... I've opted for changing the
    layer and argument names to 'drop path' rather than mix DropConnect as a layer name and use 'survival rate' as the
    argument.
    r   r   r   )r   )rD   device)rG   ndimr&   randrD   r   floor_div)r   r   r   	keep_probrG   random_tensorr   s          r7   	drop_pathr   :  s     CxII[[^

Q 77E

5ELL YYMYYy!M1FMr8   c                   x     e Zd ZdZd	dee   ddf fdZdej                  dej                  fdZ	de
fdZ xZS )
Dinov2DropPathzXDrop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).Nr   r!   c                 0    t         |           || _        y rf   )r#   r$   r   )r5   r   r6   s     r7   r$   zDinov2DropPath.__init__R  s    "r8   r   c                 D    t        || j                  | j                        S rf   )r   r   r   )r5   r   s     r7   re   zDinov2DropPath.forwardV  s    FFr8   c                 8    dj                  | j                        S )Nzp={})formatr   r5   s    r7   
extra_reprzDinov2DropPath.extra_reprY  s    }}T^^,,r8   rf   )rg   rh   ri   rj   r   floatr$   r&   rk   re   strr   rm   rn   s   @r7   r   r   O  sG    b#(5/ #T #GU\\ Gell G-C -r8   r   c                   X     e Zd Zd fdZdej
                  dej
                  fdZ xZS )	Dinov2MLPr!   c                 ~   t         |           |j                  x}}t        |j                  |j                  z        }t        j                  ||d      | _        t        |j                  t              rt        |j                     | _        n|j                  | _        t        j                  ||d      | _        y )NTr   )r#   r$   r(   rl   	mlp_ratior
   r   fc1ru   
hidden_actr   r   
activationfc2r5   r    in_featuresout_featureshidden_featuresr6   s        r7   r$   zDinov2MLP.__init__^  s    %+%7%77lf0063C3CCD99[/Ef''-$V%6%67DO$//DO99_lFr8   r   c                 l    | j                  |      }| j                  |      }| j                  |      }|S rf   )r   r   r   r   s     r7   re   zDinov2MLP.forwardi  s2    xx-|4xx-r8   r   r   rn   s   @r7   r   r   ]  s$    	GELL U\\ r8   r   c                   X     e Zd Zd fdZdej
                  dej
                  fdZ xZS )Dinov2SwiGLUFFNr!   c                 0   t         |           |j                  x}}t        |j                  |j                  z        }t        |dz  dz        dz   dz  dz  }t        j                  |d|z  d      | _        t        j                  ||d      | _        y )Nr>   r         Tr   )	r#   r$   r(   rl   r   r
   r   
weights_inweights_outr   s        r7   r$   zDinov2SwiGLUFFN.__init__q  s    %+%7%77lf0063C3CCD2Q67!;AAE))K_1D4P99_lNr8   r   c                     | j                  |      }|j                  dd      \  }}t        j                  j	                  |      |z  }| j                  |      S )Nr>   r=   rE   )r   chunkr
   rL   silur  )r5   r   x1x2hiddens        r7   re   zDinov2SwiGLUFFN.forwardz  sS    |4##A2#.B##B'",''r8   r   r   rn   s   @r7   r   r   p  s$    O(ELL (U\\ (r8   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 )Dinov2LayerzCThis corresponds to the Block class in the original implementation.r    r!   Nc                    t         |           t        j                  |j                  |j
                        | _        t        |      | _        t        |      | _
        |j                  dkD  rt        |j                        nt        j                         | _        t        j                  |j                  |j
                        | _        |j                   rt#        |      | _        nt'        |      | _        t        |      | _        y )Nepsr   )r#   r$   r
   	LayerNormr(   layer_norm_epsnorm1r   r   r   layer_scale1drop_path_rater   Identityr   norm2use_swiglu_ffnr   mlpr   layer_scale2r   s     r7   r$   zDinov2Layer.__init__  s    \\&"4"4&:O:OP
(0,V4BHBWBWZ]B](=(=>cecncncp\\&"4"4&:O:OP
  &v.DH (DH,V4r8   r   r   r   c                 D   | j                  | j                  |      ||      }|d   }| j                  |      }|dd  }| j                  |      |z   }| j	                  |      }| j                  |      }| j                  |      }| j                  |      |z   }|f|z   }|S )N)r   r   r   )r   r  r  r   r  r  r  )r5   r   r   r   self_attention_outputsr   r   layer_outputs           r7   re   zDinov2Layer.forward  s     "&JJ}%/ "0 "

 2!4,,-=>(, '78=H zz-0xx-((6 ~~l3mC/G+r8   r   )rg   rh   ri   rj   r   r$   r&   rk   r   r   r	   r   re   rm   rn   s   @r7   r	  r	    s~    M5| 5 5& -1"'	|| ELL)  	
 
uU\\5<</0%2EE	Fr8   r	  c                        e Zd Zdeddf fdZ	 	 	 	 ddej                  deej                     deded	ede	e
ef   fd
Z xZS )Dinov2Encoderr    r!   Nc                     t         |           || _        t        j                  t        |j                        D cg c]  }t        |       c}      | _        d| _	        y c c}w r   )
r#   r$   r    r
   
ModuleListrangenum_hidden_layersr	  layergradient_checkpointingr5   r    rc   r6   s      r7   r$   zDinov2Encoder.__init__  sN    ]]vG_G_A`#aAK$7#ab
&+# $bs   A#r   r   r   output_hidden_statesreturn_dictc                 t   |rdnd }|rdnd }t        | j                        D ]h  \  }}	|r||fz   }|||   nd }
| j                  r+| j                  r| j	                  |	j
                  ||
|      }n
 |	||
|      }|d   }|s`||d   fz   }j |r||fz   }|st        d |||fD              S t        |||      S )N r   r   c              3   &   K   | ]	  }||  y wrf   r&  ).0vs     r7   	<genexpr>z(Dinov2Encoder.forward.<locals>.<genexpr>  s     mq_`_lms   )last_hidden_stater   
attentions)	enumerater   r!  r   _gradient_checkpointing_func__call__tupler   )r5   r   r   r   r#  r$  all_hidden_statesall_self_attentionsilayer_modulelayer_head_masklayer_outputss               r7   re   zDinov2Encoder.forward  s     #7BD$5b4(4 	POA|#$58H$H!.7.CilO**t}} $ A A ))!#%	! !-]OM^ _)!,M &9]1=M<O&O#'	P*   1]4D Dm]4EGZ$[mmm++*
 	
r8   )NFFT)rg   rh   ri   r   r$   r&   rk   r   r   r	   r0  r   re   rm   rn   s   @r7   r  r    sz    ,| , , -1"'%* )
||)
 ELL))
  	)

 #)
 )
 
uo%	&)
r8   r  c                       e Zd ZeZdZdZdZdgZdZ	dZ
deej                  ej                  ej                  f   ddfdZy)	Dinov2PreTrainedModeldinov2rZ   Tr   r   r!   Nc                 H   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$              rnt        j                  j                  |j&                  j                  j                  t        j                        d| j                  j                        j                  |j&                  j                        |j&                  _        t        j                  j                  |j(                  j                  j                  t        j                        d| j                  j                        j                  |j(                  j                        |j(                  _        | j                  j*                  r%|j,                  j                  j                          yyt        |t.              r:|j0                  j                  j#                  | j                  j2                         yy)zInitialize the weightsr   )meanstdNg      ?)ru   r
   r   ry   inittrunc_normal_r^   datarN   r&   rO   r    initializer_rangerD   r   zero_r  fill_r   r0   r)   r*   r,   r   r   r   )r5   r   s     r7   _init_weightsz#Dinov2PreTrainedModel._init_weights  s!   fryy"))45 "$!6!6""%%emm43DKKDaDa "7 "b$$% MM {{&  &&( '-KK""$MM$$S) 01.0gg.C.C**//225==AKK11 /D / b++112	 &&+ %'GG$9$9  %%((7KK11 %: % b!!''(	 ! {{))!!&&,,. * 01NN%%dkk&B&BC 2r8   )rg   rh   ri   r   config_classbase_model_prefixmain_input_namesupports_gradient_checkpointing_no_split_modules_supports_sdpa_supports_flash_attn_2r	   r
   r   ry   r  rC  r&  r8   r7   r8  r8    s[    L $O&*#*+N!DE"))RYY*L$M DRV Dr8   r8  c                        e Zd Zdef fdZdefdZdeee	e   f   ddfdZ
e	 	 	 	 	 	 ddeej                     d	eej                     d
eej                     dee   dee   dee   deeef   fd       Z xZS )Dinov2Modelr    c                     t         |   |       || _        t        |      | _        t        |      | _        t        j                  |j                  |j                        | _        | j                          y )Nr  )r#   r$   r    r   r9   r  encoderr
   r  r(   r  	layernorm	post_initr   s     r7   r$   zDinov2Model.__init__  sY     *62$V,f&8&8f>S>ST 	r8   r!   c                 .    | j                   j                  S rf   r9   r.   r   s    r7   get_input_embeddingsz Dinov2Model.get_input_embeddings       ///r8   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)itemsrN  r   r   r   )r5   rU  r   r   s       r7   _prune_headszDinov2Model._prune_heads#  sE    
 +002 	CLE5LLu%//;;EB	Cr8   rZ   r[   r   r   r#  r$  c                    ||n| j                   j                  }||n| j                   j                  }||n| j                   j                  }|t	        d      | j                  || j                   j                        }| j                  ||      }| j                  |||||      }|d   }	| j                  |	      }	|	dddddf   }
|s|	|
f}||dd z   S t        |	|
|j                  |j                        S )z
        bool_masked_pos (`torch.BoolTensor` of shape `(batch_size, sequence_length)`):
            Boolean masked positions. Indicates which patches are masked (1) and which aren't (0). Only relevant for
            pre-training.
        Nz You have to specify pixel_values)r[   r   r   r#  r$  r   r   )r+  pooler_outputr   r,  )r    r   r#  use_return_dictr|   get_head_maskr  r9   rN  rO  r   r   r,  )r5   rZ   r[   r   r   r#  r$  embedding_outputencoder_outputssequence_outputpooled_outputhead_outputss               r7   re   zDinov2Model.forward+  s%    2C1N-TXT_T_TqTq$8$D $++JjJj 	 &1%<k$++B]B]?@@ &&y$++2O2OP	??<?Y,,/!5# ' 
 *!,..9'1a0+];L/!""555)-')77&11	
 	
r8   NNNNNN)rg   rh   ri   r   r$   r-   rS  r   rl   r   rX  r   r   r&   rk   r   r	   r   r   re   rm   rn   s   @r7   rL  rL    s    
| 
0&; 0C4T#Y+? CD C  0426,0,0/3&*4
u||,4
 "%,,/4
 ELL)	4

 $D>4
 'tn4
 d^4
 
u00	14
 4
r8   rL  z
    Dinov2 Model transformer with an image classification head on top (a linear layer on top of the final hidden state
    of the [CLS] token) e.g. for ImageNet.
    )custom_introc                        e Zd Zdeddf fdZe	 	 	 	 	 	 ddeej                     deej                     deej                     dee	   d	ee	   d
ee	   de
eef   fd       Z xZS )Dinov2ForImageClassificationr    r!   Nc                 0   t         |   |       |j                  | _        t        |      | _        |j                  dkD  r-t        j                  |j                  dz  |j                        nt        j                         | _	        | j                          y )Nr   r>   )r#   r$   
num_labelsrL  r9  r
   r   r(   r  
classifierrP  r   s     r7   r$   z%Dinov2ForImageClassification.__init__j  sy      ++!&) EKDUDUXYDYBIIf((1,f.?.?@_a_j_j_l 	
 	r8   rZ   r   labelsr   r#  r$  c                    ||n| j                   j                  }| j                  |||||      }|d   }|dddf   }	|ddddf   }
t        j                  |	|
j                  d      gd      }| j                  |      }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).
        NrZ  r   r   rE   
regressionsingle_label_classificationmulti_label_classificationr=   r>   )losslogitsr   r,  )r    r\  r9  r&   rQ   r;  ri  rN   r   problem_typerh  rD   longrl   r   squeezer   rP   r   r   r   r,  )r5   rZ   r   rj  r   r#  r$  r   r`  r)   patch_tokenslinear_inputrp  ro  loss_fctr   s                   r7   re   z$Dinov2ForImageClassification.forwardx  s!     &1%<k$++B]B]++/!5#  
 "!*#AqD)	&q!"u-yy)\->->1->-E!FAN.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$!//))	
 	
r8   rc  )rg   rh   ri   r   r$   r   r   r&   rk   r   r	   r0  r   re   rm   rn   s   @r7   rf  rf  c  s    |    04,0)-,0/3&*D
u||,D
 ELL)D
 &	D

 $D>D
 'tnD
 d^D
 
u++	,D
 D
r8   rf  zO
    Dinov2 backbone, to be used with frameworks like DETR and MaskFormer.
    c                   ~     e Zd Z fdZdefdZe	 	 	 d	dej                  de	e
   de	e
   de	e
   def
d       Z xZS )
Dinov2Backbonec                 v   t         |   |       t         | 	  |       t        |j                  dz         D cg c]  }|j
                   c}| _        t        |      | _        t        |      | _
        t        j                  |j
                  |j                        | _        | j                          y c c}w )Nr   r  )r#   r$   _init_backboner  r  r(   num_featuresr   r9   r  rN  r
   r  r  rO  rP  r"  s      r7   r$   zDinov2Backbone.__init__  s     v&9>v?W?WZ[?[9\]AV//]*62$V,f&8&8f>S>ST 	 ^s   B6r!   c                 .    | j                   j                  S rf   rR  r   s    r7   rS  z#Dinov2Backbone.get_input_embeddings  rT  r8   rZ   r#  r   r$  c                 b   ||n| j                   j                  }||n| j                   j                  }||n| j                   j                  }| j	                  |      }| j                  |d||      }|r|j                  n|d   }d}t        | j                  |      D ]  \  }	}
|	| j                  v s| j                   j                  r| j                  |
      }
| j                   j                  rn|
ddddf   }
|j                  \  }}}}| j                   j                  }|
j                  |||z  ||z  d      }
|
j!                  dddd	      j#                         }
||
fz  } |s|r|f|dd z   }|S |f|d	d z   }|S t%        ||r|j                  nd|r|j&                  
      S d
      S )a%  
        Examples:

        ```python
        >>> from transformers import AutoImageProcessor, AutoBackbone
        >>> 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)

        >>> processor = AutoImageProcessor.from_pretrained("facebook/dinov2-base")
        >>> model = AutoBackbone.from_pretrained(
        ...     "facebook/dinov2-base", out_features=["stage2", "stage5", "stage8", "stage11"]
        ... )

        >>> inputs = processor(image, return_tensors="pt")

        >>> outputs = model(**inputs)
        >>> feature_maps = outputs.feature_maps
        >>> list(feature_maps[-1].shape)
        [1, 768, 16, 16]
        ```NT)r#  r   r$  r   r&  r=   r   r   r>   )feature_mapsr   r,  )r    r\  r#  r   r9   rN  r   zipstage_namesr   apply_layernormrO  reshape_hidden_statesrG   r4   rJ   rK   r   r   r,  )r5   rZ   r#  r   r$  r^  r   r   r~  stager   rb   rc   r:   r;   r4   r   s                    r7   re   zDinov2Backbone.forward  s   @ &1%<k$++B]B]$8$D $++JjJj 	 2C1N-TXT_T_TqTq??<8,,4K\ju  
 2=--'!*#&t'7'7#G 	0E<)));;..#'>>,#?L;;44#/12#6L 4@3E3E0J65!%!7!7J#/#7#7
FjDXZ_cmZmoq#rL#/#7#71a#C#N#N#PL/	0 #&712;6 M '712;6M%3G'//T->w))
 	
 EI
 	
r8   )NNN)rg   rh   ri   r$   r-   rS  r   r&   rk   r   r   r   re   rm   rn   s   @r7   rx  rx    s{    0&; 0  04,0&*G
llG
 'tnG
 $D>	G

 d^G
 
G
 G
r8   rx  )rf  rL  r8  rx  )r   )r   F)@rj   collections.abcrv   typingr   r   r   r   r   r   r	   r&   torch.utils.checkpointr
   torch.nnr   r   r   activationsr   modeling_outputsr   r   r   r   modeling_utilsr   r   pytorch_utilsr   r   utilsr   r   r   utils.backbone_utilsr   configuration_dinov2r   
get_loggerrg   r   Moduler   r-   rk   r   r   r   r   r   r   r   r   r   r   r   r	  r  r8  rL  rf  rx  __all__r&  r8   r7   <module>r     s:     D D D    A A ! r r F Q 7 7 1 . 
		H	%Nryy NbBII R %II%<<% 
% <<	%
 U\\*% % %>;")) ;~ryy &$bii $N+ryy +U\\ e T V[VbVb *-RYY -		 &(bii ("0")) 0h0
BII 0
f &DO &D &DR M
' M
 M
` T
#8 T
T
n 
Y
*M Y

Y
x er8   