
    rh                     &   d Z ddlZddlm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 ddl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mZmZ ddlmZ  ej>                  e       Z!e ed       G d de                    Z" G d dejF                        Z$ G d dejF                        Z% G d dejF                        Z& G d dejF                        Z'	 d=dejF                  de	jP                  de	jP                  de	jP                  dee	jP                     d e)d!e)fd"Z* G d# d$ejF                        Z+ G d% d&ejF                        Z, G d' d(ejF                        Z- G d) d*ejF                        Z. G d+ d,ejF                        Z/ G d- d.e      Z0 G d/ d0ejF                        Z1e G d1 d2e             Z2e G d3 d4e2             Z3 G d5 d6ejF                        Z4 G d7 d8ejF                        Z5 ed9       G d: d;e2             Z6g d<Z7y)>zPyTorch YOLOS model.    N)	dataclass)CallableOptionalUnion)nn   )ACT2FN)GradientCheckpointingLayer)BaseModelOutputBaseModelOutputWithPooling)ALL_ATTENTION_FUNCTIONSPreTrainedModel) find_pruneable_heads_and_indicesprune_linear_layer)ModelOutputauto_docstringlogging   )YolosConfigz5
    Output type of [`YolosForObjectDetection`].
    )custom_introc                   <   e Zd ZU dZdZeej                     ed<   dZ	ee
   ed<   dZeej                     ed<   dZeej                     ed<   dZeee
      ed<   dZeej                     ed<   dZeeej                        ed	<   dZeeej                        ed
<   y)YolosObjectDetectionOutputa0  
    loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` are provided)):
        Total loss as a linear combination of a negative log-likehood (cross-entropy) for class prediction and a
        bounding box loss. The latter is defined as a linear combination of the L1 loss and the generalized
        scale-invariant IoU loss.
    loss_dict (`Dict`, *optional*):
        A dictionary containing the individual losses. Useful for logging.
    logits (`torch.FloatTensor` of shape `(batch_size, num_queries, num_classes + 1)`):
        Classification logits (including no-object) for all queries.
    pred_boxes (`torch.FloatTensor` of shape `(batch_size, num_queries, 4)`):
        Normalized boxes coordinates for all queries, represented as (center_x, center_y, width, height). These
        values are normalized in [0, 1], relative to the size of each individual image in the batch (disregarding
        possible padding). You can use [`~YolosImageProcessor.post_process`] to retrieve the unnormalized bounding
        boxes.
    auxiliary_outputs (`list[Dict]`, *optional*):
        Optional, only returned when auxiliary losses are activated (i.e. `config.auxiliary_loss` is set to `True`)
        and labels are provided. It is a list of dictionaries containing the two above keys (`logits` and
        `pred_boxes`) for each decoder layer.
    last_hidden_state (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
        Sequence of hidden-states at the output of the last layer of the decoder of the model.
    Nloss	loss_dictlogits
pred_boxesauxiliary_outputslast_hidden_statehidden_states
attentions)__name__
__module____qualname____doc__r   r   torchFloatTensor__annotations__r   dictr   r   r   listr   r   tupler         {/var/www/html/ai-insurance-compliance-backend/venv/lib/python3.12/site-packages/transformers/models/yolos/modeling_yolos.pyr   r   %   s    , )-D(5$$
%, $Ix~$*.FHU&&'..2J**+2.2xT
+259x 1 1298<M8E%"3"345<59Ju00129r,   r   c                   d     e Zd ZdZdeddf fdZdej                  dej                  fdZ xZ	S )YolosEmbeddingszT
    Construct the CLS token, detection tokens, position and patch embeddings.

    configreturnNc                 n   t         |           t        j                  t	        j
                  dd|j                              | _        t        j                  t	        j
                  d|j                  |j                              | _	        t        |      | _        | j                  j                  }t        j                  t	        j
                  d||j                  z   dz   |j                              | _        t        j                  |j                        | _        t#        |      | _        || _        y Nr   )super__init__r   	Parameterr%   zeroshidden_size	cls_tokennum_detection_tokensdetection_tokensYolosPatchEmbeddingspatch_embeddingsnum_patchesposition_embeddingsDropouthidden_dropout_probdropout$InterpolateInitialPositionEmbeddingsinterpolationr0   )selfr0   r>   	__class__s      r-   r5   zYolosEmbeddings.__init__R   s    ekk!Q8J8J&KL "U[[F<W<WY_YkYk-l m 4V <++77#%<<KK;)D)DDqH&J\J\]$
  zz&"<"<=A&Ir,   pixel_valuesc                    |j                   \  }}}}| j                  |      }|j                         \  }}}| j                  j	                  |dd      }	| j
                  j	                  |dd      }
t        j                  |	||
fd      }| j                  | j                  ||f      }||z   }| j                  |      }|S )Nr   dim)shaper=   sizer9   expandr;   r%   catrD   r?   rB   )rE   rG   
batch_sizenum_channelsheightwidth
embeddingsseq_len_
cls_tokensr;   r?   s               r-   forwardzYolosEmbeddings.forwarda   s    2>2D2D/
L&%**<8
!+!2
GQ ^^**:r2>
0077
BKYY
J8HIqQ
 #001I1IFTY?["55
\\*-
r,   
r!   r"   r#   r$   r   r5   r%   TensorrX   __classcell__rF   s   @r-   r/   r/   L   s6    
{ t ELL U\\ r,   r/   c                   B     e Zd Zd fdZddej
                  fdZ xZS )rC   r1   c                 0    t         |           || _        y Nr4   r5   r0   rE   r0   rF   s     r-   r5   z-InterpolateInitialPositionEmbeddings.__init__x       r,   c                    |d d dd d f   }|d d d f   }|d d | j                   j                   d d d f   }|d d d| j                   j                   d d f   }|j                  dd      }|j                  \  }}}| j                   j                  d   | j                   j
                  z  | j                   j                  d   | j                   j
                  z  }
}	|j                  |||	|
      }|\  }}|| j                   j
                  z  || j                   j
                  z  }}t        j                  j                  |||fdd      }|j                  d      j                  dd      }t        j                  |||fd      }|S )Nr   r      bicubicFrM   modealign_cornersrJ   )r0   r:   	transposerL   
image_size
patch_sizeviewr   
functionalinterpolateflattenr%   rO   )rE   	pos_embedimg_sizecls_pos_embeddet_pos_embedpatch_pos_embedrP   r8   rU   patch_heightpatch_widthrR   rS   new_patch_heightnew_patch_widthscale_pos_embeds                   r-   rX   z,InterpolateInitialPositionEmbeddings.forward|   s   !!Q'*%ag.!!dkk&F&F%F%H!"KL#AqDKK,L,L+L'La$OP)33Aq9+:+@+@(
K KK""1%)?)??KK""1%)?)?? " *..z;Vab ,2dkk6L6L,LeW[WbWbWmWmNm/--33#3_"EIej 4 
 *11!4>>q!D))]O]$SYZ[r,   r1   N)i   i@  r!   r"   r#   r5   r%   rZ   rX   r[   r\   s   @r-   rC   rC   w   s    %,, r,   rC   c                   B     e Zd Zd fdZddej
                  fdZ xZS ) InterpolateMidPositionEmbeddingsr1   c                 0    t         |           || _        y r_   r`   ra   s     r-   r5   z)InterpolateMidPositionEmbeddings.__init__   rb   r,   c                 v   |d d d d dd d f   }|d d d f   }|d d d d | j                   j                   d d d f   }|d d d d d| j                   j                   d d f   }|j                  dd      }|j                  \  }}}}	| j                   j                  d   | j                   j
                  z  | j                   j                  d   | j                   j
                  z  }}
|j                  ||z  ||
|      }|\  }}|| j                   j
                  z  || j                   j
                  z  }}t        j                  j                  |||fdd      }|j                  d      j                  dd      j                         j                  ||||z  |      }t        j                  |||fd      }|S )	Nr   r   rd   r   re   Frf   rJ   )r0   r:   ri   rL   rj   rk   rl   r   rm   rn   ro   
contiguousr%   rO   )rE   rp   rq   rr   rs   rt   depthrP   r8   rU   ru   rv   rR   rS   rw   rx   ry   s                    r-   rX   z(InterpolateMidPositionEmbeddings.forward   s   !!Q1*-%ag.!!Q)I)I(I(KQ"NO#Aq!t{{/O/O.O*OQR$RS)33Aq92A2G2G/z; KK""1%)?)??KK""1%)?)?? " *..uz/A;P\^ij ,2dkk6L6L,LeW[WbWbWmWmNm/--33#3_"EIej 4 
 ##A&Yq!_Z\T%%5%GU	 	  ))]O]$SYZ[r,   rz   r{   r|   r\   s   @r-   r~   r~      s    %,, r,   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)r4   r5   rj   rk   rQ   r8   
isinstancecollectionsabcIterabler>   r   Conv2d
projection)rE   r0   rj   rk   rQ   r8   r>   rF   s          r-   r5   zYolosPatchEmbeddings.__init__   s    !'!2!2F4E4EJ
$*$7$79K9Kk#-j+//:R:R#SZZdfpYq
#-j+//:R:R#SZZdfpYq
!!}
15*Q-:VW=:XY$$(&))L+:^hir,   rG   r1   c                     |j                   \  }}}}|| j                  k7  rt        d      | j                  |      j	                  d      j                  dd      }|S )NzeMake sure that the channel dimension of the pixel values match with the one set in the configuration.rd   r   )rL   rQ   
ValueErrorr   ro   ri   )rE   rG   rP   rQ   rR   rS   rT   s          r-   rX   zYolosPatchEmbeddings.forward   sb    2>2D2D/
L&%4,,,w  __\2::1=GG1M
r,   )	r!   r"   r#   r$   r5   r%   rZ   rX   r[   r\   s   @r-   r<   r<      s)    jELL U\\ r,   r<   modulequerykeyvalueattention_maskscalingrB   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 )NrI   )rK   dtype)ptrainingr   rd   )r%   matmulri   r   rm   softmaxfloat32tor   rB   r   r   )
r   r   r   r   r   r   rB   kwargsattn_weightsattn_outputs
             r-   eager_attention_forwardr      s     <<s}}R'<=GL ==((2U]](SVVW\WbWbcL ==((6??([L !#n4,,|U3K''1-88:K$$r,   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 )	YolosSelfAttentionr0   r1   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 .g      F)bias)r4   r5   r8   num_attention_headshasattrr   r0   intattention_head_sizeall_head_sizeattention_probs_dropout_probdropout_probr   	is_causalr   Linearqkv_biasr   r   r   ra   s     r-   r5   zYolosSelfAttention.__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\
r,   	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 )
NrI   r   rd   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   rB   r   )rL   r   rl   r   r   ri   r   r   r   r0   _attn_implementationloggerwarning_oncer   r   r   r   r   rM   r   reshape)rE   r   r   r   rP   
seq_lengthrV   	key_layervalue_layerquery_layerattention_interfacecontext_layerattention_probsnew_context_layer_shapeoutputss                  r-   rX   zYolosSelfAttention.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$2C2C	*
& #0"4"4"6s";t?Q?Q>S"S%--.EF6G=/2 O\M]r,   NF)r!   r"   r#   r   r5   r   r%   rZ   boolr   r*   rX   r[   r\   s   @r-   r   r      sr    ]{ ]t ]. -1"'	1 ELL)1  	1
 
uU\\5<</0%2EE	F1r,   r   c                   |     e Zd ZdZdeddf fdZdej                  dej                  dej                  fdZ xZ	S )	YolosSelfOutputz
    The residual connection is defined in YolosLayer instead of here (as is the case with other models), due to the
    layernorm applied before each block.
    r0   r1   Nc                     t         |           t        j                  |j                  |j                        | _        t        j                  |j                        | _        y r_   )	r4   r5   r   r   r8   denser@   rA   rB   ra   s     r-   r5   zYolosSelfOutput.__init__F  sB    YYv1163E3EF
zz&"<"<=r,   r   input_tensorc                 J    | j                  |      }| j                  |      }|S r_   r   rB   rE   r   r   s      r-   rX   zYolosSelfOutput.forwardK  s$    

=1]3r,   rY   r\   s   @r-   r   r   @  sD    
>{ >t >
U\\  RWR^R^ r,   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 )YolosAttentionr0   r1   Nc                     t         |           t        |      | _        t	        |      | _        t               | _        y r_   )r4   r5   r   	attentionr   outputsetpruned_headsra   s     r-   r5   zYolosAttention.__init__T  s0    +F3%f-Er,   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   rJ   )lenr   r   r   r   r   r   r   r   r   r   r   r   union)rE   r   indexs      r-   prune_headszYolosAttention.prune_headsZ  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:r,   r   r   r   c                 h    | j                  |||      }| j                  |d   |      }|f|dd  z   }|S )Nr   r   )r   r   )rE   r   r   r   self_outputsattention_outputr   s          r-   rX   zYolosAttention.forwardl  sE     ~~mY@QR;;|AF#%QR(88r,   r   )r!   r"   r#   r   r5   r   r   r   r%   rZ   r   r   r   r*   rX   r[   r\   s   @r-   r   r   S  s    "{ "t ";S ;d ;* -1"'	|| ELL)  	
 
uU\\5<</0%2EE	Fr,   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 )YolosIntermediater0   r1   Nc                    t         |           t        j                  |j                  |j
                        | _        t        |j                  t              rt        |j                     | _        y |j                  | _        y r_   )r4   r5   r   r   r8   intermediate_sizer   r   
hidden_actstrr	   intermediate_act_fnra   s     r-   r5   zYolosIntermediate.__init__|  s]    YYv1163K3KL
f''-'-f.?.?'@D$'-'8'8D$r,   r   c                 J    | j                  |      }| j                  |      }|S r_   )r   r   )rE   r   s     r-   rX   zYolosIntermediate.forward  s&    

=100?r,   	r!   r"   r#   r   r5   r%   rZ   rX   r[   r\   s   @r-   r   r   {  s1    9{ 9t 9U\\ ell r,   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 )YolosOutputr0   r1   Nc                     t         |           t        j                  |j                  |j
                        | _        t        j                  |j                        | _	        y r_   )
r4   r5   r   r   r   r8   r   r@   rA   rB   ra   s     r-   r5   zYolosOutput.__init__  sB    YYv779K9KL
zz&"<"<=r,   r   r   c                 T    | j                  |      }| j                  |      }||z   }|S r_   r   r   s      r-   rX   zYolosOutput.forward  s.    

=1]3%4r,   r   r\   s   @r-   r   r     s?    >{ >t >
U\\  RWR^R^ r,   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 )
YolosLayerz?This corresponds to the Block class in the timm implementation.r0   r1   Nc                 r   t         |           |j                  | _        d| _        t	        |      | _        t        |      | _        t        |      | _	        t        j                  |j                  |j                        | _        t        j                  |j                  |j                        | _        y )Nr   eps)r4   r5   chunk_size_feed_forwardseq_len_dimr   r   r   intermediater   r   r   	LayerNormr8   layer_norm_epslayernorm_beforelayernorm_afterra   s     r-   r5   zYolosLayer.__init__  s    '-'E'E$'/-f5!&) "V-?-?VEZEZ [!||F,>,>FDYDYZr,   r   r   r   c                     | j                  | j                  |      ||      }|d   }|dd  }||z   }| j                  |      }| j                  |      }| j	                  ||      }|f|z   }|S )N)r   r   r   )r   r   r   r   r   )rE   r   r   r   self_attention_outputsr   r   layer_outputs           r-   rX   zYolosLayer.forward  s     "&!!-0/ "0 "

 2!4(, )=8 ++M:((6 {{<?/G+r,   r   )r!   r"   r#   r$   r   r5   r%   rZ   r   r   r   r*   rX   r[   r\   s   @r-   r   r     s    I[{ [t [ -1"'	|| ELL)  	
 
uU\\5<</0%2EE	Fr,   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 )YolosEncoderr0   r1   Nc                 @   t         |           || _        t        j                  t        |j                        D cg c]  }t        |       c}      | _        d| _	        d|j                  d   |j                  d   z  |j                  dz  z  z   |j                  z   }|j                  rBt        j                  t        j                   |j                  dz
  d||j"                              nd | _        |j                  rt'        |      | _        y d | _        y c c}w )NFr   r   rd   )r4   r5   r0   r   
ModuleListrangenum_hidden_layersr   layergradient_checkpointingrj   rk   r:   use_mid_position_embeddingsr6   r%   r7   r8   mid_position_embeddingsr~   rD   )rE   r0   rV   r   rF   s       r-   r5   zYolosEncoder.__init__  s   ]]fF^F^@_#`1Jv$6#`a
&+# ""1%(9(9!(<<@Q@QST@TTUX^XsXss 	 11 LL,,q0&&	  	$ JPIkIk=fEqu' $as   Dr   r   r   output_hidden_statesreturn_dictc                    |rdnd }|rdnd }	| j                   j                  r| j                  | j                  ||f      }
t	        | j
                        D ]k  \  }}|r||fz   }|||   nd } ||||      }|d   }| j                   j                  r$|| j                   j                  dz
  k  r|
|   z   }|sc|	|d   fz   }	m |r||fz   }|st        d |||	fD              S t        |||	      S )Nr+   r   r   c              3   &   K   | ]	  }||  y wr_   r+   ).0vs     r-   	<genexpr>z'YolosEncoder.forward.<locals>.<genexpr>  s     mq_`_lms   )r   r   r    )	r0   r  rD   r  	enumerater  r   r*   r   )rE   r   rR   rS   r   r   r  r  all_hidden_statesall_self_attentions$interpolated_mid_position_embeddingsilayer_modulelayer_head_masklayer_outputss                  r-   rX   zYolosEncoder.forward  s.    #7BD$5b4;;22373E3EdFbFbekmrds3t0(4 	POA|#$58H$H!.7.CilO(IZ[M)!,M{{66559:$14XYZ4[$[M &9]1=M<O&O#	P"   1]4D Dm]4EGZ$[mmm++*
 	
r,   )NFFT)r!   r"   r#   r   r5   r%   rZ   r   r   r   r*   r   rX   r[   r\   s   @r-   r   r     s}    v{ vt v: -1"'%* *
||*

 ELL)*
  *
 #*
 *
 
uo%	&*
r,   r   c                       e Zd ZU eed<   dZdZdZ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)	YolosPreTrainedModelr0   vitrG   Tr   r1   Nc                    t        |t        j                  t        j                  f      rm|j                  j
                  j                  d| j                  j                         |j                  %|j                  j
                  j                          yyt        |t        j                        rJ|j                  j
                  j                          |j                  j
                  j                  d       yy)zInitialize the weightsr   )meanstdNg      ?)r   r   r   r   weightdatanormal_r0   initializer_ranger   zero_r   fill_)rE   r   s     r-   _init_weightsz"YolosPreTrainedModel._init_weights  s    fryy"))45 MM&&CT[[5R5R&S{{&  &&( '-KK""$MM$$S) .r,   )r!   r"   r#   r   r'   base_model_prefixmain_input_namesupports_gradient_checkpointing_no_split_modules_supports_sdpa_supports_flash_attn_supports_flex_attn_supports_attention_backendr   r   r   r   r   r   r+   r,   r-   r  r    sb    $O&*#N"&
*E"))RYY*L$M 
*RV 
*r,   r  c                        e Zd Zd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   dee   dee   deeef   fd       Z xZS )
YolosModelr0   add_pooling_layerc                    t         |   |       || _        t        |      | _        t        |      | _        t        j                  |j                  |j                        | _        |rt        |      nd| _        | j                          y)zv
        add_pooling_layer (bool, *optional*, defaults to `True`):
            Whether to add a pooling layer
        r   N)r4   r5   r0   r/   rT   r   encoderr   r   r8   r   	layernormYolosPoolerpooler	post_init)rE   r0   r+  rF   s      r-   r5   zYolosModel.__init__'  sk    
 	 )&1#F+f&8&8f>S>ST->k&)D 	r,   r1   c                 .    | j                   j                  S r_   )rT   r=   )rE   s    r-   get_input_embeddingszYolosModel.get_input_embeddings8  s    ///r,   heads_to_pruneNc                     |j                         D ]7  \  }}| j                  j                  |   j                  j	                  |       9 y)a	  
        Prunes heads of the model.

        Args:
            heads_to_prune (`dict`):
                See base class `PreTrainedModel`. The input dictionary must have the following format: {layer_num:
                list of heads to prune in this layer}
        N)itemsr-  r  r   r   )rE   r4  r  r   s       r-   _prune_headszYolosModel._prune_heads;  sE     +002 	CLE5LLu%//;;EB	Cr,   rG   r   r   r  r  c           	      `   ||n| j                   j                  }||n| j                   j                  }||n| j                   j                  }|t	        d      | j                  || j                   j                        }| j                  |      }| j                  ||j                  d   |j                  d   ||||      }|d   }| j                  |      }| j                  | j                  |      nd }	|s|	||	fn|f}
|
|dd  z   S t        ||	|j                  |j                        S )Nz You have to specify pixel_valuesr   rI   )rR   rS   r   r   r  r  r   r   )r   pooler_outputr   r    )r0   r   r  use_return_dictr   get_head_maskr   rT   r-  rL   r.  r0  r   r   r    )rE   rG   r   r   r  r  embedding_outputencoder_outputssequence_outputpooled_outputhead_outputss              r-   rX   zYolosModel.forwardG  sR    2C1N-TXT_T_TqTq$8$D $++JjJj 	 &1%<k$++B]B]?@@ &&y$++2O2OP	??<8,,%%b)$$R(/!5# ' 
 *!,..98<8OO4UY?L?XO];_n^pL/!""555)-')77&11	
 	
r,   )T)NNNNN)r!   r"   r#   r   r   r5   r<   r3  r(   r   r)   r7  r   r   r%   rZ   r   r*   r   rX   r[   r\   s   @r-   r*  r*  %  s    { t "0&: 0
C4T#Y+? 
CD 
C  04,0,0/3&*0
u||,0
 ELL)0
 $D>	0

 'tn0
 d^0
 
u00	10
 0
r,   r*  c                   *     e Zd Zdef fdZd Z xZS )r/  r0   c                     t         |           t        j                  |j                  |j                        | _        t        j                         | _        y r_   )r4   r5   r   r   r8   r   Tanh
activationra   s     r-   r5   zYolosPooler.__init__|  s9    YYv1163E3EF
'')r,   c                 \    |d d df   }| j                  |      }| j                  |      }|S )Nr   )r   rD  )rE   r   first_token_tensorr?  s       r-   rX   zYolosPooler.forward  s6     +1a40

#566r,   )r!   r"   r#   r   r5   rX   r[   r\   s   @r-   r/  r/  {  s    ${ $
r,   r/  c                   (     e Zd ZdZ fdZd Z xZS )YolosMLPPredictionHeada  
    Very simple multi-layer perceptron (MLP, also called FFN), used to predict the normalized center coordinates,
    height and width of a bounding box w.r.t. an image.

    Copied from https://github.com/facebookresearch/detr/blob/master/models/detr.py

    c                     t         |           || _        |g|dz
  z  }t        j                  d t        |g|z   ||gz         D              | _        y )Nr   c              3   N   K   | ]  \  }}t        j                  ||        y wr_   )r   r   )r	  nks      r-   r  z2YolosMLPPredictionHead.__init__.<locals>.<genexpr>  s     #g1BIIaO#gs   #%)r4   r5   
num_layersr   r   ziplayers)rE   	input_dim
hidden_dim
output_dimrM  hrF   s         r-   r5   zYolosMLPPredictionHead.__init__  sS    $LJN+mm#gYKRSOUVZdYeUe@f#ggr,   c                     t        | j                        D ]D  \  }}|| j                  dz
  k  r%t        j                  j                   ||            n ||      }F |S r3   )r  rO  rM  r   rm   relu)rE   xr  r  s       r-   rX   zYolosMLPPredictionHead.forward  sT    !$++. 	VHAu01DOOa4G0G""58,USTXA	Vr,   )r!   r"   r#   r$   r5   rX   r[   r\   s   @r-   rH  rH    s    hr,   rH  zy
    YOLOS Model (consisting of a ViT encoder) with object detection heads on top, for tasks such as COCO detection.
    c                        e Zd Zdef fdZej                  j                  d        Ze		 	 	 	 ddej                  deee      dee   dee   dee   d	eeef   fd
       Z xZS )YolosForObjectDetectionr0   c                 "   t         |   |       t        |d      | _        t	        |j
                  |j
                  |j                  dz   d      | _        t	        |j
                  |j
                  dd      | _        | j                          y )NF)r+  r   r   )rP  rQ  rR  rM     )
r4   r5   r*  r  rH  r8   
num_labelsclass_labels_classifierbbox_predictorr1  ra   s     r-   r5   z YolosForObjectDetection.__init__  s      f> (>((V5G5GTZTeTehiTivw(
$ 5((V5G5GTUbc

 	r,   c                 ^    t        |d d |d d       D cg c]
  \  }}||d c}}S c c}}w )NrI   )r   r   )rN  )rE   outputs_classoutputs_coordabs        r-   _set_aux_lossz%YolosForObjectDetection._set_aux_loss  s9    
 <?}Sb?QS`adbdSe;fg41a1A.gggs   )rG   labelsr   r  r  r1   c           
         ||n| j                   j                  }| j                  ||||      }|d   }|dd| j                   j                   dddf   }| j	                  |      }| j                  |      j                         }	d\  }
}}|d\  }}| j                   j                  rC|r|j                  n|d   }| j	                  |      }| j                  |      j                         }| j                  ||| j                  |	| j                   ||      \  }
}}|s|||	f|z   |z   }n||	f|z   }|
|
|f|z   S |S t        |
|||	||j                  |j                  |j                        S )a	  
        labels (`list[Dict]` of len `(batch_size,)`, *optional*):
            Labels for computing the bipartite matching loss. List of dicts, each dictionary containing at least the
            following 2 keys: `'class_labels'` and `'boxes'` (the class labels and bounding boxes of an image in the
            batch respectively). The class labels themselves should be a `torch.LongTensor` of len `(number of bounding
            boxes in the image,)` and the boxes a `torch.FloatTensor` of shape `(number of bounding boxes in the image,
            4)`.

        Examples:

        ```python
        >>> from transformers import AutoImageProcessor, AutoModelForObjectDetection
        >>> 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("hustvl/yolos-tiny")
        >>> model = AutoModelForObjectDetection.from_pretrained("hustvl/yolos-tiny")

        >>> inputs = image_processor(images=image, return_tensors="pt")
        >>> outputs = model(**inputs)

        >>> # convert outputs (bounding boxes and class logits) to Pascal VOC format (xmin, ymin, xmax, ymax)
        >>> target_sizes = torch.tensor([image.size[::-1]])
        >>> results = image_processor.post_process_object_detection(outputs, threshold=0.9, target_sizes=target_sizes)[
        ...     0
        ... ]

        >>> for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
        ...     box = [round(i, 2) for i in box.tolist()]
        ...     print(
        ...         f"Detected {model.config.id2label[label.item()]} with confidence "
        ...         f"{round(score.item(), 3)} at location {box}"
        ...     )
        Detected remote with confidence 0.991 at location [46.48, 72.78, 178.98, 119.3]
        Detected remote with confidence 0.908 at location [336.48, 79.27, 368.23, 192.36]
        Detected cat with confidence 0.934 at location [337.18, 18.06, 638.14, 373.09]
        Detected cat with confidence 0.979 at location [10.93, 53.74, 313.41, 470.67]
        Detected remote with confidence 0.974 at location [41.63, 72.23, 178.09, 119.99]
        ```N)r   r  r  r   )NNN)NNrZ  )r   r   r   r   r   r   r   r    )r0   r:  r  r:   r\  r]  sigmoidauxiliary_lossintermediate_hidden_statesloss_functiondevicer   r   r   r    )rE   rG   rd  r   r  r  r   r>  r   r   r   r   r   r_  r`  r   r   s                    r-   rX   zYolosForObjectDetection.forward  s   h &1%<k$++B]B] ((/!5#	  
 "!* *!dkk.N.N-N-PRS*ST --o>((9AAC
-=*i*+5(M={{))EPwAAV]^_V` $ < <\ J $ 3 3L A I I K151C1CZmUb2.D).  , *-0AAGK *-7373CT9%.OO)!/%77!//))	
 		
r,   )NNNN)r!   r"   r#   r   r5   r%   jitunusedrc  r   r&   r   r)   r(   r   r   r*   r   rX   r[   r\   s   @r-   rX  rX    s    { & YYh h  (,,0/3&*a
''a
 d$a
 $D>	a

 'tna
 d^a
 
u00	1a
 a
r,   rX  )rX  r*  r  )r   )8r$   collections.abcr   dataclassesr   typingr   r   r   r%   torch.utils.checkpointr   activationsr	   modeling_layersr
   modeling_outputsr   r   modeling_utilsr   r   pytorch_utilsr   r   utilsr   r   r   configuration_yolosr   
get_loggerr!   r   r   Moduler/   rC   r~   r<   rZ   floatr   r   r   r   r   r   r   r   r  r*  r/  rH  rX  __all__r+   r,   r-   <module>r|     s#     ! , ,    ! 9 K F Q 9 9 , 
		H	% 
: : :B(bii (V299 :ryy B299 R %II%<<% 
% <<	%
 U\\*% % %>F FTbii &$RYY $P		 ""))  '+ 'TC
299 C
L *? * *0 R
% R
 R
j"))  RYY * 
}
2 }

}
@ Lr,   