
    Z j              	          S r SSKrSSKrSSKJr  SSKrSSKJrJr  SSK	J
r  SSKJr  SSKJrJr  SS	KJr  SS
KJr  SSKJr  SSKJrJrJrJr  SSKJr  SSKJr  \R@                  " \!5      r"\\" SS9 " S S\5      5       5       r#\\" SS9 " S S\5      5       5       r$\\" SS9 " S S\5      5       5       r%\\" SS9 " S S\5      5       5       r&S r'S r(SJS\R                  S \)S!\*S"\R                  4S# jjr+ " S$ S%\RX                  5      r- " S& S'\RX                  5      r. " S( S)\RX                  5      r/ " S* S+\RX                  5      r0 " S, S-\RX                  5      r1 " S. S/\RX                  5      r2 " S0 S1\RX                  5      r3 " S2 S3\RX                  5      r4 " S4 S5\RX                  5      r5 " S6 S7\RX                  5      r6 " S8 S9\5      r7 " S: S;\RX                  5      r8\ " S< S=\5      5       r9\ " S> S?\95      5       r:\" S@S9 " SA SB\95      5       r;\" SCS9 " SD SE\95      5       r<\" SFS9 " SG SH\\95      5       r=/ SIQr>g)Kz!PyTorch Swinv2 Transformer model.    N)	dataclass)Tensornn   )initialization)ACT2FN)BackboneMixinfilter_output_hidden_states)GradientCheckpointingLayer)BackboneOutput)PreTrainedModel)ModelOutputauto_docstringlogging	torch_int)can_return_tuple   )Swinv2ConfigzP
    Swinv2 encoder's outputs, with potential hidden states and attentions.
    )custom_introc                       \ rS rSr% SrSr\R                  S-  \S'   Sr	\
\R                  S4   S-  \S'   Sr\
\R                  S4   S-  \S'   Sr\
\R                  S4   S-  \S'   S	rg)
Swinv2EncoderOutput(   a  
reshaped_hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
    Tuple of `torch.FloatTensor` (one for the output of the embeddings + one for the output of each stage) of
    shape `(batch_size, hidden_size, height, width)`.

    Hidden-states of the model at the output of each layer plus the initial embedding outputs reshaped to
    include the spatial dimensions.
Nlast_hidden_state.hidden_states
attentionsreshaped_hidden_states )__name__
__module____qualname____firstlineno____doc__r   torchFloatTensor__annotations__r   tupler   r   __static_attributes__r       {/root/GenerationalWealth/GenerationalWealth/venv/lib/python3.13/site-packages/transformers/models/swinv2/modeling_swinv2.pyr   r   (   s}     37u((4/6:>M5**C/047>7;Je'',-4;CGE%"3"3S"89D@Gr(   r   zX
    Swinv2 model's outputs that also contains a pooling of the last hidden states.
    c                       \ rS rSr% SrSr\R                  S-  \S'   Sr	\R                  S-  \S'   Sr
\\R                  S4   S-  \S'   Sr\\R                  S4   S-  \S'   Sr\\R                  S4   S-  \S	'   S
rg)Swinv2ModelOutput?   a  
pooler_output (`torch.FloatTensor` of shape `(batch_size, hidden_size)`, *optional*, returned when `add_pooling_layer=True` is passed):
    Average pooling of the last layer hidden-state.
reshaped_hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
    Tuple of `torch.FloatTensor` (one for the output of the embeddings + one for the output of each stage) of
    shape `(batch_size, hidden_size, height, width)`.

    Hidden-states of the model at the output of each layer plus the initial embedding outputs reshaped to
    include the spatial dimensions.
Nr   pooler_output.r   r   r   r   )r   r   r    r!   r"   r   r#   r$   r%   r-   r   r&   r   r   r'   r   r(   r)   r+   r+   ?   s    	 37u((4/6.2M5$$t+2:>M5**C/047>7;Je'',-4;CGE%"3"3S"89D@Gr(   r+   z,
    Swinv2 masked image model outputs.
    c                       \ rS rSr% SrSr\R                  S-  \S'   Sr	\R                  S-  \S'   Sr
\\R                  S4   S-  \S'   Sr\\R                  S4   S-  \S'   Sr\\R                  S4   S-  \S	'   S
rg)Swinv2MaskedImageModelingOutputY   a  
loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `bool_masked_pos` is provided):
    Masked image modeling (MLM) loss.
reconstruction (`torch.FloatTensor` of shape `(batch_size, num_channels, height, width)`):
    Reconstructed pixel values.
reshaped_hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
    Tuple of `torch.FloatTensor` (one for the output of the embeddings + one for the output of each stage) of
    shape `(batch_size, hidden_size, height, width)`.

    Hidden-states of the model at the output of each layer plus the initial embedding outputs reshaped to
    include the spatial dimensions.
Nlossreconstruction.r   r   r   r   )r   r   r    r!   r"   r1   r#   r$   r%   r2   r   r&   r   r   r'   r   r(   r)   r/   r/   Y   s     &*D%

d
")/3NE%%,3:>M5**C/047>7;Je'',-4;CGE%"3"3S"89D@Gr(   r/   z2
    Swinv2 outputs for image classification.
    c                       \ rS rSr% SrSr\R                  S-  \S'   Sr	\R                  S-  \S'   Sr
\\R                  S4   S-  \S'   Sr\\R                  S4   S-  \S'   Sr\\R                  S4   S-  \S	'   S
rg)Swinv2ImageClassifierOutputu   a  
loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided):
    Classification (or regression if config.num_labels==1) loss.
logits (`torch.FloatTensor` of shape `(batch_size, config.num_labels)`):
    Classification (or regression if config.num_labels==1) scores (before SoftMax).
reshaped_hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
    Tuple of `torch.FloatTensor` (one for the output of the embeddings + one for the output of each stage) of
    shape `(batch_size, hidden_size, height, width)`.

    Hidden-states of the model at the output of each layer plus the initial embedding outputs reshaped to
    include the spatial dimensions.
Nr1   logits.r   r   r   r   )r   r   r    r!   r"   r1   r#   r$   r%   r6   r   r&   r   r   r'   r   r(   r)   r4   r4   u   s     &*D%

d
")'+FE$+:>M5**C/047>7;Je'',-4;CGE%"3"3S"89D@Gr(   r4   c                     U R                   u  p#pEU R                  X#U-  XU-  X5      n U R                  SSSSSS5      R                  5       R                  SXU5      nU$ )z*
Partitions the given input into windows.
r   r   r            shapeviewpermute
contiguous)input_featurewindow_size
batch_sizeheightwidthnum_channelswindowss          r)   window_partitionrH      so     /<.A.A+J!&&k);8LkM ##Aq!Q15@@BGGKfrsGNr(   c                     U R                   S   nU R                  SX!-  X1-  XU5      n U R                  SSSSSS5      R                  5       R                  SX#U5      n U $ )z7
Merges windows to produce higher resolution features.
r;   r   r   r   r8   r9   r:   r<   )rG   rB   rD   rE   rF   s        r)   window_reverserJ      se     ==$Lll2v4e6JKfrsGooaAq!Q/::<AA"fUabGNr(   input	drop_probtrainingreturnc                    US:X  d  U(       d  U $ SU-
  nU R                   S   4SU R                  S-
  -  -   nU[        R                  " X@R                  U R
                  S9-   nUR                  5         U R                  U5      U-  nU$ )z[
Drop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).

        r   r   )r   )dtypedevice)r=   ndimr#   randrQ   rR   floor_div)rK   rL   rM   	keep_probr=   random_tensoroutputs          r)   	drop_pathrZ      s    
 CxII[[^

Q 77E

5ELL YYMYYy!M1FMr(   c                      ^  \ rS rSrSrSS\S-  SS4U 4S jjjrS\R                  S\R                  4S jr	S\
4S	 jrS
rU =r$ )Swinv2DropPath   zXDrop paths (Stochastic Depth) per sample (when applied in main path of residual blocks).NrL   rN   c                 .   > [         TU ]  5         Xl        g N)super__init__rL   )selfrL   	__class__s     r)   ra   Swinv2DropPath.__init__   s    "r(   r   c                 B    [        XR                  U R                  5      $ r_   )rZ   rL   rM   rb   r   s     r)   forwardSwinv2DropPath.forward   s    FFr(   c                      SU R                    3$ )Nzp=rL   rb   s    r)   
extra_reprSwinv2DropPath.extra_repr   s    DNN#$$r(   rj   r_   )r   r   r    r!   r"   floatra   r#   r   rg   strrl   r'   __classcell__rc   s   @r)   r\   r\      sQ    b#%$, #$ # #GU\\ Gell G%C % %r(   r\   c            
          ^  \ rS rSrSrSU 4S jjrS\R                  S\S\S\R                  4S jr	  SS
\R                  S	-  S\R                  S	-  S\S\\R                     4S jjrSrU =r$ )Swinv2Embeddings   zO
Construct the patch and position embeddings. Optionally, also the mask token.
c                   > [         TU ]  5         [        U5      U l        U R                  R                  nU R                  R
                  U l        U(       a6  [        R                  " [        R                  " SSUR                  5      5      OS U l        UR                  (       a?  [        R                  " [        R                  " SUS-   UR                  5      5      U l        OS U l        [        R                  " UR                  5      U l        [        R"                  " UR$                  5      U l        UR(                  U l        Xl        g )Nr   )r`   ra   Swinv2PatchEmbeddingspatch_embeddingsnum_patches	grid_size
patch_gridr   	Parameterr#   zeros	embed_dim
mask_tokenuse_absolute_embeddingsposition_embeddings	LayerNormnormDropouthidden_dropout_probdropout
patch_sizeconfig)rb   r   use_mask_tokenrx   rc   s       r)   ra   Swinv2Embeddings.__init__   s     5f =++77//99O]",,u{{1a9I9I'JKcg))')||EKK;QR?TZTdTd4e'fD$'+D$LL!1!12	zz&"<"<= ++r(   
embeddingsrD   rE   rN   c                    UR                   S   S-
  nU R                  R                   S   S-
  n[        R                  R	                  5       (       d  XE:X  a  X#:X  a  U R                  $ U R                  SS2SS24   nU R                  SS2SS24   nUR                   S   nX R
                  -  n	X0R
                  -  n
[        US-  5      nUR                  SXU5      nUR                  SSSS5      n[        R                  R                  UX4SS	S
9nUR                  SSSS5      R                  SSU5      n[        R                  " Xg4SS9$ )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   Nr;         ?r   r   r8   bicubicF)sizemodealign_cornersdim)r=   r   r#   jit
is_tracingr   r   reshaper?   r   
functionalinterpolater>   cat)rb   r   rD   rE   rx   num_positionsclass_pos_embedpatch_pos_embedr   
new_height	new_widthsqrt_num_positionss               r)   interpolate_pos_encoding)Swinv2Embeddings.interpolate_pos_encoding   sS    !&&q)A-0066q9A= yy##%%+*F6?+++221bqb59221ab59r".
__,	&}c'9:)11!5G]`a)11!Q1=--33(	 4 
 *11!Q1=BB1b#Nyy/;CCr(   Npixel_valuesbool_masked_posr   c                    UR                   u  pEpgU R                  U5      u  pU R                  U5      nUR                  5       u  pnUbI  U R                  R                  XS5      nUR                  S5      R                  U5      nUSU-
  -  X-  -   nU R                  b*  U(       a  XR                  XU5      -   nOXR                  -   nU R                  U5      nX4$ )Nr;         ?)r=   rw   r   r   r~   expand	unsqueezetype_asr   r   r   )rb   r   r   r   _rF   rD   rE   r   output_dimensionsrC   seq_lenmask_tokensmasks                 r)   rg   Swinv2Embeddings.forward	  s     *6););&(,(=(=l(K%
YYz*
!+!2
Q&//00bIK",,R088ED#sTz2[5GGJ##/''*G*G
\a*bb
'*B*BB
\\*-
,,r(   )r   r   r~   r   rw   rz   r   r   FNF)r   r   r    r!   r"   ra   r#   r   intr   r$   
BoolTensorboolr&   rg   r'   rp   rq   s   @r)   rs   rs      s    &&D5<< &D &DUX &D]b]i]i &DV 48).	-''$.- ))D0- #'	-
 
u||	- -r(   rs   c                      ^  \ rS rSrSrU 4S jrS rS\R                  S-  S\	\R                  \	\   4   4S jrS	rU =r$ )
rv   i&  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U ]  5         UR                  UR                  p2UR                  UR
                  pT[        U[        R                  R                  5      (       a  UOX"4n[        U[        R                  R                  5      (       a  UOX34nUS   US   -  US   US   -  -  nX l        X0l        X@l        X`l
        US   US   -  US   US   -  4U l        [        R                  " XEX3S9U l        g )Nr   r   )kernel_sizestride)r`   ra   
image_sizer   rF   r}   
isinstancecollectionsabcIterablerx   ry   r   Conv2d
projection)rb   r   r   r   rF   hidden_sizerx   rc   s          r)   ra   Swinv2PatchEmbeddings.__init__-  s    !'!2!2F4E4EJ$*$7$79I9Ik#-j+//:R:R#S#SZZdYq
#-j+//:R:R#S#SZZdYq
!!}
15*Q-:VW=:XY$$(&$Q-:a=8*Q-:VW=:XY))L:ir(   c                 f   X0R                   S   -  S:w  aB  SU R                   S   X0R                   S   -  -
  4n[        R                  R                  X5      nX R                   S   -  S:w  aD  SSSU R                   S   X R                   S   -  -
  4n[        R                  R                  X5      nU$ )Nr   r   )r   r   r   pad)rb   r   rD   rE   
pad_valuess        r)   	maybe_padSwinv2PatchEmbeddings.maybe_pad<  s    ??1%%*T__Q/%//!:L2LLMJ==,,\FLOOA&&!+Q4??1#5QRAS8S#STJ==,,\FLr(   r   NrN   c                     UR                   u  p#pEU R                  XU5      nU R                  U5      nUR                   u    p$nXE4nUR                  S5      R	                  SS5      nXg4$ )Nr8   r   )r=   r   r   flatten	transpose)rb   r   r   rF   rD   rE   r   r   s           r)   rg   Swinv2PatchEmbeddings.forwardE  sp    )5););&~~lEB__\2
(..1e#O''*44Q:
,,r(   )ry   r   rF   rx   r   r   )r   r   r    r!   r"   ra   r   r#   r$   r&   r   r   rg   r'   rp   rq   s   @r)   rv   rv   &  sK    j	-E$5$5$< 	-u||UZ[^U_G_A` 	- 	-r(   rv   c            	          ^  \ rS rSrSr\R                  4S\\   S\S\R                  SS4U 4S jjjr
S	 rS
\R                  S\\\4   S\R                  4S jrSrU =r$ )Swinv2PatchMergingiQ  a  
Patch Merging Layer.

Args:
    input_resolution (`tuple[int]`):
        Resolution of input feature.
    dim (`int`):
        Number of input channels.
    norm_layer (`nn.Module`, *optional*, defaults to `nn.LayerNorm`):
        Normalization layer class.
input_resolutionr   
norm_layerrN   Nc                    > [         TU ]  5         Xl        X l        [        R
                  " SU-  SU-  SS9U l        U" SU-  5      U l        g )Nr9   r8   Fbias)r`   ra   r   r   r   Linear	reductionr   )rb   r   r   r   rc   s       r)   ra   Swinv2PatchMerging.__init__^  sE     01s7AG%@q3w'	r(   c                     US-  S:H  =(       d    US-  S:H  nU(       a-  SSSUS-  SUS-  4n[         R                  R                  X5      nU$ )Nr8   r   r   )r   r   r   )rb   rA   rD   rE   
should_padr   s         r)   r   Swinv2PatchMerging.maybe_pade  sS    qjAo:519>
Q519a!<JMM--mHMr(   rA   input_dimensionsc                    Uu  p4UR                   u  pVnUR                  XSXG5      nU R                  XU5      nUS S 2SS S2SS S2S S 24   nUS S 2SS S2SS S2S S 24   n	US S 2SS S2SS S2S S 24   n
US S 2SS S2SS S2S S 24   n[        R                  " XX/S5      nUR                  USSU-  5      nU R                  U5      nU R                  U5      nU$ )Nr   r8   r   r;   r9   )r=   r>   r   r#   r   r   r   )rb   rA   r   rD   rE   rC   r   rF   input_feature_0input_feature_1input_feature_2input_feature_3s               r)   rg   Swinv2PatchMerging.forwardm  s   ((5(;(;%
%**:uS}eD'14a4Aq(89'14a4Aq(89'14a4Aq(89'14a4Aq(89		?_"fhjk%**:r1|;KL}5		-0r(   )r   r   r   r   )r   r   r    r!   r"   r   r   r&   r   Modulera   r   r#   r   rg   r'   rp   rq   s   @r)   r   r   Q  s|    
 XZWcWc (s (# (299 (hl ( (U\\ U3PS8_ Y^YeYe  r(   r   c            
          ^  \ rS rSrSS/4U 4S jjr  SS\R                  S\R                  S-  S\S-  S\	\R                     4S	 jjr
S
 rSrU =r$ )Swinv2SelfAttentioni  r   c           
      h  > [         TU ]  5         X#-  S:w  a  [        SU SU S35      eX0l        [	        X#-  5      U l        U R                  U R
                  -  U l        [        U[        R                  R                  5      (       a  UOXD4U l        XPl        [        R                  " [        R                   " S[        R"                  " USS45      -  5      5      U l        [        R&                  " [        R(                  " SSS	S
9[        R*                  " S	S9[        R(                  " SUSS
95      U l        U R/                  5       u  pgU R1                  SUSS9  U R1                  SUSS9  [        R(                  " U R                  U R                  UR2                  S
9U l        [        R(                  " U R                  U R                  SS
9U l        [        R(                  " U R                  U R                  UR2                  S
9U l        [        R:                  " UR<                  5      U l        g )Nr   zThe hidden size (z6) is not a multiple of the number of attention heads ()
   r   r8   i   Tr   )inplaceFrelative_coords_table)
persistentrelative_position_index) r`   ra   
ValueErrornum_attention_headsr   attention_head_sizeall_head_sizer   r   r   r   rB   pretrained_window_sizer   r{   r#   logoneslogit_scale
Sequentialr   ReLUcontinuous_position_bias_mlpcreate_coords_table_and_indexregister_bufferqkv_biasquerykeyvaluer   attention_probs_dropout_probr   )	rb   r   r   	num_headsrB   r   r   r   rc   s	           r)   ra   Swinv2SelfAttention.__init__  s   ?a#C5(^_h^iijk  $- #&s#7 !558P8PP%k;??3K3KLLKS^Rl 	 '=#<<		"uzz9aQRBS7T2T(UV,.MMIIa4("''$*?3PY`eAf-
) :>9[9[9]646KX]^68O\abYYt1143E3EFOO\
99T//1C1C%PYYt1143E3EFOO\
zz&"E"EFr(   Nr   attention_maskoutput_attentionsrN   c                 :   UR                   u  pEnU R                  U5      R                  USU R                  U R                  5      R                  SS5      nU R                  U5      R                  USU R                  U R                  5      R                  SS5      nU R                  U5      R                  USU R                  U R                  5      R                  SS5      n	[        R                  R                  USS9[        R                  R                  USS9R                  SS5      -  n
[        R                  " U R                  [        R                  " S5      S9R!                  5       nX-  n
U R#                  U R$                  5      R                  SU R                  5      nXR&                  R                  S5         R                  U R(                  S   U R(                  S   -  U R(                  S   U R(                  S   -  S5      nUR+                  SSS5      R-                  5       nS	[        R.                  " U5      -  nXR1                  S5      -   n
Ub  UR                   S   nU
R                  XN-  XR                  XU5      UR1                  S5      R1                  S5      -   n
XR1                  S5      R1                  S5      -   n
U
R                  SU R                  XU5      n
[        R                  R3                  U
SS9nU R5                  U5      n[        R6                  " X5      nUR+                  SSSS
5      R-                  5       nUR9                  5       S S U R:                  4-   nUR                  U5      nU(       a  UU4nU$ U4nU$ )Nr;   r   r8   r   g      Y@)maxr      r   )r=   r   r>   r   r   r   r   r   r   r   	normalizer#   clampr   mathr   expr   r   r   rB   r?   r@   sigmoidr   softmaxr   matmulr   r   )rb   r   r   r   rC   r   rF   query_layer	key_layervalue_layerattention_scoresr   relative_position_bias_tablerelative_position_bias
mask_shapeattention_probscontext_layernew_context_layer_shapeoutputss                      r)   rg   Swinv2SelfAttention.forward  sa    )6(;(;%
JJ}%T*b$":":D<T<TUYq!_ 	 HH]#T*b$":":D<T<TUYq!_ 	 JJ}%T*b$":":D<T<TUYq!_ 	 ==22;B2G"--JaJa2 Kb K

)B
 kk$"2"28LMQQS+9'+'H'HIcIc'd'i'i(((
$ ">>Z>Z>_>_`b>c!d!i!iQ$"2"21"55t7G7G7JTM]M]^_M`7`bd"
 "8!?!?1a!H!S!S!U!#emm4J&K!K+.N.Nq.QQ%'--a0J/44(*6N6NPS ((+55a8 9  02J2J12M2W2WXY2ZZ/44R9Q9QSV\ --//0@b/I ,,7 _B%--aAq9DDF"/"4"4"6s";t?Q?Q>S"S%**+BC6G=/2 O\M]r(   c                 8   [         R                  " U R                  S   S-
  * U R                  S   [         R                  S9R	                  5       n[         R                  " U R                  S   S-
  * U R                  S   [         R                  S9R	                  5       n[         R
                  " [         R                  " X/SS95      R                  SSS5      R                  5       R                  S5      nU R                  S   S:  aO  US S 2S S 2S S 2S4==   U R                  S   S-
  -  ss'   US S 2S S 2S S 2S4==   U R                  S   S-
  -  ss'   OaU R                  S   S:  aN  US S 2S S 2S S 2S4==   U R                  S   S-
  -  ss'   US S 2S S 2S S 2S4==   U R                  S   S-
  -  ss'   US-  n[         R                  " U5      [         R                  " [         R                  " U5      S-   5      -  [        R                  " S5      -  nUR                  [!        U R"                  R%                  5       5      R&                  5      n[         R                  " U R                  S   5      n[         R                  " U R                  S   5      n[         R
                  " [         R                  " XE/SS95      n[         R(                  " US5      nUS S 2S S 2S 4   US S 2S S S 24   -
  nUR                  SSS5      R                  5       nUS S 2S S 2S4==   U R                  S   S-
  -  ss'   US S 2S S 2S4==   U R                  S   S-
  -  ss'   US S 2S S 2S4==   SU R                  S   -  S-
  -  ss'   UR+                  S	5      n	X94$ )
Nr   r   rQ   ij)indexingr8      r   r;   )r#   arangerB   int64rn   stackmeshgridr?   r@   r   r   signlog2absr  tonextr   
parametersrQ   r   sum)
rb   relative_coords_hrelative_coords_wr   coords_hcoords_wcoordscoords_flattenrelative_coordsr   s
             r)   r   1Swinv2SelfAttention.create_coords_table_and_index  s7   !LL4+;+;A+>+B)CTEUEUVWEX`e`k`klrrt!LL4+;+;A+>+B)CTEUEUVWEX`e`k`klrrtKK(9'MX\]^WQ1Z\Yq\	 	 &&q)A-!!Q1*-1L1LQ1ORS1SS-!!Q1*-1L1LQ1ORS1SS-a 1$!!Q1*-1A1A!1Dq1HH-!!Q1*-1A1A!1Dq1HH-"JJ,-

599EZ;[^a;a0bbeienenopeqq 	 !6 8 8d>_>_>j>j>l9m9s9s t << 0 0 34<< 0 0 34U^^X,@4PQvq1(At4~aqj7QQ)11!Q:EEG1a D$4$4Q$7!$;; 1a D$4$4Q$7!$;; 1a A(8(8(;$;a$?? "1"5"5b"9$==r(   )r   r   r   r   r   r   r   r   r   r   rB   r   )r   r   r    r!   ra   r#   r   r$   r   r&   rg   r   r'   rp   rq   s   @r)   r   r     sr    TUWXSY G@ 48).	B||B ))D0B  $;	B
 
u||	BH#> #>r(   r   c                   z   ^  \ rS rSrU 4S jrS\R                  S\R                  S\R                  4S jrSrU =r	$ )Swinv2SelfOutputi  c                    > [         TU ]  5         [        R                  " X"5      U l        [        R
                  " UR                  5      U l        g r_   )r`   ra   r   r   denser   r   r   rb   r   r   rc   s      r)   ra   Swinv2SelfOutput.__init__  s4    YYs(
zz&"E"EFr(   r   input_tensorrN   c                 J    U R                  U5      nU R                  U5      nU$ r_   r/  r   )rb   r   r2  s      r)   rg   Swinv2SelfOutput.forward  s$    

=1]3r(   r4  
r   r   r    r!   ra   r#   r   rg   r'   rp   rq   s   @r)   r-  r-    s7    G
U\\  RWR^R^  r(   r-  c            
          ^  \ rS rSrS
U 4S jjr  SS\R                  S\R                  S-  S\S-  S\	\R                     4S jjr
S	rU =r$ )Swinv2Attentioni  c           
         > [         TU ]  5         [        UUUU[        U[        R
                  R                  5      (       a  UOXU4S9U l        [        X5      U l	        g )Nr   r   r   rB   r   )
r`   ra   r   r   r   r   r   rb   r-  rY   )rb   r   r   r   rB   r   rc   s         r)   ra   Swinv2Attention.__init__  sW    '#0+//2J2JKK $:(A
	 'v3r(   Nr   r   r   rN   c                 f    U R                  XU5      nU R                  US   U5      nU4USS  -   nU$ )Nr   r   )rb   rY   )rb   r   r   r   self_outputsattention_outputr  s          r)   rg   Swinv2Attention.forward+  sC     yy@QR;;|AF#%QR(88r(   )rY   rb   r   r   )r   r   r    r!   ra   r#   r   r$   r   r&   rg   r'   rp   rq   s   @r)   r8  r8    s\    4  48).		||	 ))D0	  $;		
 
u||		 	r(   r8  c                   b   ^  \ rS rSrU 4S jrS\R                  S\R                  4S jrSrU =r	$ )Swinv2Intermediatei8  c                   > [         TU ]  5         [        R                  " U[	        UR
                  U-  5      5      U l        [        UR                  [        5      (       a  [        UR                     U l        g UR                  U l        g r_   )r`   ra   r   r   r   	mlp_ratior/  r   
hidden_actro   r   intermediate_act_fnr0  s      r)   ra   Swinv2Intermediate.__init__9  sd    YYsC(8(83(>$?@
f''--'-f.?.?'@D$'-'8'8D$r(   r   rN   c                 J    U R                  U5      nU R                  U5      nU$ r_   r/  rF  rf   s     r)   rg   Swinv2Intermediate.forwardA  s&    

=100?r(   rI  r6  rq   s   @r)   rB  rB  8  s(    9U\\ ell  r(   rB  c                   b   ^  \ rS rSrU 4S jrS\R                  S\R                  4S jrSrU =r	$ )Swinv2OutputiH  c                    > [         TU ]  5         [        R                  " [	        UR
                  U-  5      U5      U l        [        R                  " UR                  5      U l	        g r_   )
r`   ra   r   r   r   rD  r/  r   r   r   r0  s      r)   ra   Swinv2Output.__init__I  sF    YYs6#3#3c#9:C@
zz&"<"<=r(   r   rN   c                 J    U R                  U5      nU R                  U5      nU$ r_   r4  rf   s     r)   rg   Swinv2Output.forwardN  s$    

=1]3r(   r4  r6  rq   s   @r)   rL  rL  H  s(    >
U\\ ell  r(   rL  c                      ^  \ rS rSr SU 4S jjrS\\\\4   \\\4   4   4S jrS rS r	 SS\
R                  S\\\4   S	\S
-  S\\
R                  \
R                  4   4S jjrSrU =r$ )Swinv2LayeriT  c           
      ^  > [         T	U ]  5         X0l        U R                  UR                  UR                  4Xf45      u  pUS   U l        US   U l        [        UUUU R                  [        U[        R                  R                  5      (       a  UOXw4S9U l        [        R                  " X!R                  S9U l        US:  a  [!        U5      O[        R"                  " 5       U l        ['        X5      U l        [+        X5      U l        [        R                  " X!R                  S9U l        g )Nr   r:  epsrP   )r`   ra   r   _compute_window_shiftrB   
shift_sizer8  r   r   r   r   	attentionr   r   layer_norm_epslayernorm_beforer\   IdentityrZ   rB  intermediaterL  rY   layernorm_after)
rb   r   r   r   r   drop_path_raterW  r   rB   rc   s
            r)   ra   Swinv2Layer.__init__U  s    	 0"&"<"<!3!34z6N#
 'q>$Q-(((0+//2J2JKK $:(A
 !#S6K6K L;IC;O7UWU`U`Ub.v;"6/!||C5J5JKr(   rN   c                     [        U R                  U5       VVs/ s H  u  p4[        X45      PM     nnn[        U R                  XR5       VVVs/ s H  u  p4ocU::  a  SOUPM     nnnnXW4$ s  snnf s  snnnf Nr   )zipr   min)rb   target_window_sizetarget_shift_sizerwrB   srW  s           r)   rV  !Swinv2Layer._compute_window_shiftn  so    -01F1FHZ-[\-[TQs1y-[\8;D<Q<QS^8rs8rWQ16aq(8r
s&& ]ss   A+A1c           	         U R                   S:  Gae  [        R                  " SXS4US9n[        SU R                  * 5      [        U R                  * U R                   * 5      [        U R                   * S 5      4n[        SU R                  * 5      [        U R                  * U R                   * 5      [        U R                   * S 5      4nSnU H  nU H  n	XtS S 2XS S 24'   US-  nM     M     [        X@R                  5      n
U
R                  SU R                  U R                  -  5      n
U
R                  S5      U
R                  S5      -
  nUR                  US:g  S5      R                  US:H  S5      nU$ S nU$ )Nr   r   r  r;   r8   g      YrP   )	rW  r#   r|   slicerB   rH   r>   r   masked_fill)rb   rD   rE   rQ   img_maskheight_sliceswidth_slicescountheight_slicewidth_slicemask_windows	attn_masks               r)   get_attn_maskSwinv2Layer.get_attn_masks  sy   ??Q{{Ava#8FHa$***+t'''$//)9:t&-M a$***+t'''$//)9:t&-L
 E -#/K@EQ1<=QJE $0 !.
 ,H6F6FGL',,R1A1ADDTDT1TUL$..q1L4J4J14MMI!--i1nfEQQR[_`R`befI  Ir(   c                     U R                   X0R                   -  -
  U R                   -  nU R                   X R                   -  -
  U R                   -  nSSSUSU4n[        R                  R                  X5      nX4$ ra  )rB   r   r   r   )rb   r   rD   rE   	pad_right
pad_bottomr   s          r)   r   Swinv2Layer.maybe_pad  sy    %%0@0@(@@DDTDTT	&&2B2B)BBdFVFVV
Ay!Z8
))-D((r(   r   r   r   Nc                    Uu  pEUR                  5       u  pgnUn	UR                  XdXX5      nU R                  XU5      u  pUR                  u  p{pU R                  S:  a.  [
        R                  " XR                  * U R                  * 4SS9nOUn[        XR                  5      nUR                  SU R                  U R                  -  U5      nU R                  XUR                  S9nUb  UR                  UR                  5      nU R                  XUS9nUS   nUR                  SU R                  U R                  U5      n[        UU R                  X5      nU R                  S:  a-  [
        R                  " UU R                  U R                  4SS9nOUnU
S   S:  =(       d    U
S   S:  nU(       a  US S 2S U2S U2S S 24   R                  5       nUR                  XdU-  U5      nU R!                  U5      nXR#                  U5      -   nU R%                  U5      nU R'                  U5      nXR#                  U R)                  U5      5      -   nU(       a	  UUS	   4nU$ U4nU$ )
Nr   )r   r8   )shiftsdimsr;   r  )r   r   r:   r   )r   r>   r   r=   rW  r#   rollrH   rB   ru  rQ   r   rR   rX  rJ   r@   rZ  rZ   r\  rY   r]  )rb   r   r   r   rD   rE   rC   r   channelsshortcutr   
height_pad	width_padshifted_hidden_stateshidden_states_windowsrt  attention_outputsr>  attention_windowsshifted_windows
was_paddedlayer_outputlayer_outputss                          r)   rg   Swinv2Layer.forward  sx    )"/"4"4"6
x  &**:uO$(NN=%$P!&3&9&9#y??Q$)JJ}FVY]YhYhXhEipv$w!$1! !11FHXHX Y 5 : :2t?O?ORVRbRb?bdl m&&zMDWDW&X	 !%:%A%ABI NN+@_pNq,Q/,11"d6F6FHXHXZbc():D<L<Ljd ??Q %

?DOOUYUdUdCelr s /]Q&;*Q-!*;
 1!WfWfufa2G H S S U-22:~xX--.?@ >>-#@@((7{{<0$~~d6J6J<6X'YY@Q'8';< YeWfr(   )	rX  rZ   r   r\  r]  rZ  rY   rW  rB   )rP   r   r   r   )r   r   r    r!   ra   r&   r   rV  ru  r   r#   r   r   rg   r'   rp   rq   s   @r)   rR  rR  T  s    qrL2'eTYZ]_bZbTcejknpsksetTtNu '
8) */	5||5  S/5  $;	5
 
u||U\\)	*5 5r(   rR  c            
          ^  \ rS rSr S
U 4S jjr SS\R                  S\\\4   S\	S-  S\\R                     4S jjr
S	rU =r$ )Swinv2Stagei  c	                 d  > [         TU ]  5         Xl        X l        / n	[	        U5       H=  n
[        UUUUXj   U
S-  S:X  a  SOUR                  S-  US9nU	R                  U5        M?     [        R                  " U	5      U l
        Ub  U" X2[        R                  S9U l        OS U l        SU l        g )Nr8   r   )r   r   r   r   r^  rW  r   )r   r   F)r`   ra   r   r   rangerR  rB   appendr   
ModuleListblocksr   
downsamplepointing)rb   r   r   r   depthr   rZ   r  r   r  iblockrc   s               r)   ra   Swinv2Stage.__init__  s     	uA!1#(|!"Q!1&2D2D2I'=E MM%   mmF+ !()9r||\DO"DOr(   r   r   r   NrN   c                     Uu  pE[        U R                  5       H  u  pgU" UUU5      nUS   nM     Un	U R                  b%  US-   S-  US-   S-  pXEX4nU R                  X5      nOXEXE4nXU4nU(       a  UWSS  -  nU$ )Nr   r   r8   )	enumerater  r  )rb   r   r   r   rD   rE   r  layer_moduler  !hidden_states_before_downsamplingheight_downsampledwidth_downsampledr   stage_outputss                 r)   rg   Swinv2Stage.forward  s     )(5OA( !M *!,M  6 -:)??&5;aZA4EPQ	VWGW 1!'0B V OO,M`M!' >&K\]]12..Mr(   )r  r   r   r  r  r@  r   )r   r   r    r!   ra   r#   r   r&   r   r   rg   r'   rp   rq   s   @r)   r  r    sZ    mn@ */	||  S/  $;	
 
u||	 r(   r  c                      ^  \ rS rSrSU 4S jjr    SS\R                  S\\\4   S\	S-  S\	S-  S\	S-  S	\	S-  S
\\
-  4S jjrSrU =r$ )Swinv2Encoderi
  c                 <  > [         T	U ]  5         [        UR                  5      U l        Xl        U R
                  R                  b  UR                  n[        R                  " SUR                  [        UR                  5      SS9 Vs/ s H  oDR                  5       PM     nn/ n[        U R                  5       H  n[        U[        UR                  SU-  -  5      US   SU-  -  US   SU-  -  4UR                  U   UR                   U   U[        UR                  S U 5      [        UR                  S US-    5       XpR                  S-
  :  a  ["        OS X7   S9nUR%                  U5        M     [&        R(                  " U5      U l        SU l        g s  snf )Nr   cpu)rR   r8   r   )r   r   r   r  r   rZ   r  r   F)r`   ra   lendepths
num_layersr   pretrained_window_sizesr#   linspacer^  r#  itemr  r  r   r}   r   r   r  r   r  layersgradient_checkpointing)
rb   r   ry   r  xdprr  i_layerstagerc   s
            r)   ra   Swinv2Encoder.__init__  sm   fmm,;;..:&,&D&D#!&63H3H#fmmJ\ej!kl!kAvvx!klT__-G((1g:56"+A,1g:">	!QRT[Q[@\!]mmG, **73c&--"9:S}QX[\Q\A]=^_29OOa<O2O-VZ'>'G	E MM%  . mmF+&+## ms   	Fr   r   r   Noutput_hidden_states(output_hidden_states_before_downsamplingreturn_dictrN   c                    U(       a  SOS nU(       a  SOS nU(       a  SOS n	U(       aB  UR                   u  pnUR                  " U
/UQUP76 nUR                  SSSS5      nXq4-  nX4-  n[        U R                  5       H  u  pU" UUU5      nUS   nUS   nUS   nUS   US   4nU(       aS  U(       aL  UR                   u  pnUR                  " U
/US   US   4QUP76 nUR                  SSSS5      nUU4-  nX4-  nOPU(       aI  U(       dB  UR                   u  pnUR                  " U
/UQUP76 nUR                  SSSS5      nXq4-  nX4-  nU(       d  M  U	USS  -  n	M     U(       d  [        S XX4 5       5      $ [        UUU	US	9$ )
Nr   r   r   r   r8   r   r;   c              3   0   #    U  H  nUc  M  Uv   M     g 7fr_   r   ).0vs     r)   	<genexpr>(Swinv2Encoder.forward.<locals>.<genexpr>\  s      lA ls   	)r   r   r   r   )r=   r>   r?   r  r  r&   r   )rb   r   r   r   r  r  r  all_hidden_statesall_reshaped_hidden_statesall_self_attentionsrC   r   r   reshaped_hidden_stater  r  r  r  r   s                      r)   rg   Swinv2Encoder.forward$  s    #7BD+?RT"$5b4)6)<)<&J;$1$6$6z$bDT$bVa$b!$9$A$A!Q1$M!!11&*BB&(5OA( !M *!,M0=a0@- -a 0 1" 57H7LM#(P-N-T-T*
{ )J(N(N)"3A"68I!8L!M)OZ)% )>(E(EaAq(Q%!&G%II!*.FF*%.V-:-@-@*
{(5(:(::(fHX(fZe(f%(=(E(EaAq(Q%!%55!*.FF*  #}QR'88#A  6D  '<Ol   #++*#=	
 	
r(   )r   r  r  r  ))r   r   r   r   )FFFT)r   r   r    r!   ra   r#   r   r&   r   r   r   rg   r'   rp   rq   s   @r)   r  r  
  s    ,: */,1@E#'C
||C
  S/C
  $;	C

 #TkC
 37+C
 D[C
 
$	$C
 C
r(   r  c                   d    \ rS rSr% \\S'   SrSrSrSr	S/r
\R                  " 5       S 5       rS	rg
)Swinv2PreTrainedModelij  r   swinv2r   )imageTr  c                    [        U[        R                  [        R                  45      (       ac  [        R
                  " UR                  SU R                  R                  S9  UR                  b!  [        R                  " UR                  5        gg[        U[        R                  5      (       aA  [        R                  " UR                  5        [        R                  " UR                  5        g[        U[        5      (       a\  UR                  b   [        R                  " UR                  5        UR                  b!  [        R                  " UR                  5        gg[        U[         5      (       a  [        R"                  " UR$                  [&        R(                  " S5      5        UR+                  5       u  p#[        R,                  " UR.                  U5        [        R,                  " UR0                  U5        gg)zInitialize the weightsrP   )meanstdNr   )r   r   r   r   initnormal_weightr   initializer_ranger   zeros_r   ones_rs   r~   r   r   	constant_r   r  r   r   copy_r   r   )rb   moduler   r   s       r)   _init_weights#Swinv2PreTrainedModel._init_weightss  sF    fryy"))455LLSdkk6S6ST{{&FKK( '--KK$JJv}}% 011  ,F--.))5F667 6 344NN6--txx|<=C=a=a=c:!JJv335JKJJv557NO	 5r(   r   N)r   r   r    r!   r   r%   base_model_prefixmain_input_nameinput_modalitiessupports_gradient_checkpointing_no_split_modulesr#   no_gradr  r'   r   r(   r)   r  r  j  sA     $O!&*#&
]]_P Pr(   r  c                      ^  \ rS rSrSU 4S jjrS r\      SS\R                  S-  S\R                  S-  S\
S-  S\
S-  S	\
S
\
S-  S\\-  4S jj5       rSrU =r$ )Swinv2Modeli  c                   > [         TU ]  U5        Xl        [        UR                  5      U l        [        UR                  SU R
                  S-
  -  -  5      U l        [        XS9U l
        [        XR                  R                  5      U l        [        R                  " U R                  UR                   S9U l        U(       a  [        R$                  " S5      OSU l        U R)                  5         g)z
add_pooling_layer (`bool`, *optional*, defaults to `True`):
    Whether or not to apply pooling layer.
use_mask_token (`bool`, *optional*, defaults to `False`):
    Whether or not to create and apply mask tokens in the embedding layer.
r8   r   )r   rT  N)r`   ra   r   r  r  r  r   r}   num_featuresrs   r   r  rz   encoderr   r   rY  	layernormAdaptiveAvgPool1dpooler	post_init)rb   r   add_pooling_layerr   rc   s       r)   ra   Swinv2Model.__init__  s     	 fmm, 0 0119L3M MN*6Q$V__-G-GHd&7&7V=R=RS1Bb**1- 	r(   c                 .    U R                   R                  $ r_   r   rw   rk   s    r)   get_input_embeddings Swinv2Model.get_input_embeddings      ///r(   Nr   r   r   r  r   r  rN   c                 >   Ub  UOU R                   R                  nUb  UOU R                   R                  nUb  UOU R                   R                  nUc  [	        S5      eU R                  XUS9u  pU R                  UU	UUUS9n
U
S   nU R                  U5      nSnU R                  b8  U R                  UR                  SS5      5      n[        R                  " US5      nU(       d  X4U
SS -   nU$ [        UUU
R                  U
R                  U
R                  S9$ )	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)r   r   )r   r  r  r   r   r8   )r   r-   r   r   r   )r   r   r  r  r   r   r  r  r  r   r#   r   r+   r   r   r   )rb   r   r   r   r  r   r  kwargsembedding_outputr   encoder_outputssequence_outputpooled_outputrY   s                 r)   rg   Swinv2Model.forward  sC    2C1N-TXT_T_TqTq$8$D $++JjJj 	 &1%<k$++BYBY?@@-1__Tl .= .
* ,,/!5# ' 
 *!,..9;;" KK(A(A!Q(GHM!MM-;M%58KKFM -')77&11#2#I#I
 	
r(   )r   r   r  r  r  r  r  )TFNNNNFN)r   r   r    r!   ra   r  r   r#   r$   r   r   r&   r+   rg   r'   rp   rq   s   @r)   r  r    s    *0  2637)-,0).#'6
''$.6
 ))D06
  $;	6

 #Tk6
 #'6
 D[6
 
"	"6
 6
r(   r  a~  
        Swinv2 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>
    c                      ^  \ rS rSrU 4S jr\      SS\R                  S-  S\R                  S-  S\	S-  S\	S-  S\	S	\	S-  S
\
\-  4S jj5       rSrU =r$ )Swinv2ForMaskedImageModelingi  c                   > [         TU ]  U5        [        USSS9U l        [	        UR
                  SUR                  S-
  -  -  5      n[        R                  " [        R                  " X!R                  S-  UR                  -  SS9[        R                  " UR                  5      5      U l        U R                  5         g )NFT)r  r   r8   r   )in_channelsout_channelsr   )r`   ra   r  r  r   r}   r  r   r   r   encoder_striderF   PixelShuffledecoderr  )rb   r   r  rc   s      r)   ra   %Swinv2ForMaskedImageModeling.__init__  s     !&ERVW6++aF4E4E4I.JJK}}II(7L7La7ORXReRe7est OOF112	
 	r(   Nr   r   r   r  r   r  rN   c           	         Ub  UOU R                   R                  nU R                  UUUUUUS9nUS   n	U	R                  SS5      n	U	R                  u  pn[
        R                  " US-  5      =pU	R                  XX5      n	U R                  U	5      nSnUGb  U R                   R                  U R                   R                  -  nUR                  SUU5      nUR                  U R                   R                  S5      R                  U R                   R                  S5      R                  S5      R                  5       n[        R                  R!                  XSS	9nUU-  R#                  5       UR#                  5       S
-   -  U R                   R$                  -  nU(       d  U4USS -   nUb  U4U-   $ U$ ['        UUUR(                  UR*                  UR,                  S9$ )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, Swinv2ForMaskedImageModeling
>>> import torch
>>> from PIL import Image
>>> import httpx
>>> from io import BytesIO

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> with httpx.stream("GET", url) as response:
...     image = Image.open(BytesIO(response.read()))

>>> image_processor = AutoImageProcessor.from_pretrained("microsoft/swinv2-tiny-patch4-window8-256")
>>> model = Swinv2ForMaskedImageModeling.from_pretrained("microsoft/swinv2-tiny-patch4-window8-256")

>>> 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, 256, 256]
```N)r   r   r  r   r  r   r   r8   r   r;   none)r   gh㈵>)r1   r2   r   r   r   )r   r  r  r   r=   r  floorr   r  r   r   repeat_interleaver   r@   r   r   l1_lossr#  rF   r/   r   r   r   )rb   r   r   r   r  r   r  r  r  r  rC   rF   sequence_lengthrD   rE   reconstructed_pixel_valuesmasked_im_lossr   r   reconstruction_lossrY   s                        r)   rg   $Swinv2ForMaskedImageModeling.forward  s   P &1%<k$++BYBY+++/!5%=#  
 "!*)33Aq94C4I4I1
/OS$899)11*FZ &*\\/%B"&;;))T[[-C-CCD-55b$EO11$++2H2H!L""4;;#9#91=1	  #%--"7"7lr"7"s1D8==?488:PTCTUX\XcXcXpXppN02WQR[@F3A3M^%.YSYY.5!//))#*#A#A
 	
r(   )r  r  r  )r   r   r    r!   ra   r   r#   r$   r   r   r&   r/   rg   r'   rp   rq   s   @r)   r  r    s       2637)-,0).#'S
''$.S
 ))D0S
  $;	S

 #TkS
 #'S
 D[S
 
0	0S
 S
r(   r  a  
    Swinv2 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 SwinV2 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                      ^  \ rS rSrU 4S jr\      SS\R                  S-  S\R                  S-  S\	S-  S\	S-  S\	S	\	S-  S
\
\-  4S jj5       rSrU =r$ )Swinv2ForImageClassificationiT  c                 D  > [         TU ]  U5        UR                  U l        [        U5      U l        UR                  S:  a5  [
        R                  " U R                  R                  UR                  5      O[
        R                  " 5       U l	        U R                  5         g ra  )r`   ra   
num_labelsr  r  r   r   r  r[  
classifierr  )rb   r   rc   s     r)   ra   %Swinv2ForImageClassification.__init__d  sx      ++!&) GMFWFWZ[F[BIIdkk..0A0ABacalalan 	
 	r(   Nr   labelsr   r  r   r  rN   c                 X   Ub  UOU R                   R                  nU R                  UUUUUS9nUS   n	U R                  U	5      n
SnUb  U R	                  X*U R                   5      nU(       d  U
4USS -   nUb  U4U-   $ U$ [        UU
UR                  UR                  UR                  S9$ )ab  
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   r  r   r8   )r1   r6   r   r   r   )	r   r  r  r  loss_functionr4   r   r   r   )rb   r   r	  r   r  r   r  r  r  r  r6   r1   rY   s                r)   rg   $Swinv2ForImageClassification.forwardr  s    " &1%<k$++BYBY++/!5%=#  
  
/%%fdkkBDY,F)-)9TGf$EvE*!//))#*#A#A
 	
r(   )r  r  r  r  )r   r   r    r!   ra   r   r#   r$   
LongTensorr   r&   r4   rg   r'   rp   rq   s   @r)   r  r  T  s       26*.)-,0).#',
''$.,
   4',
  $;	,

 #Tk,
 #',
 D[,
 
,	,,
 ,
r(   r  zO
    Swinv2 backbone, to be used with frameworks like DETR and MaskFormer.
    c                      ^  \ rS rSrU 4S jrS r\\\   SS\	S\
S-  S\
S-  S\
S-  S	\4
S
 jj5       5       5       rSrU =r$ )Swinv2Backbonei  c           	      l  > [         TU ]  U5        UR                  /[        [	        UR
                  5      5       Vs/ s H  n[        UR                  SU-  -  5      PM      sn-   U l        [        U5      U l	        [        XR                  R                  5      U l        U R                  5         g s  snf )Nr8   )r`   ra   r}   r  r  r  r   r  rs   r   r  rz   r  r  )rb   r   r  rc   s      r)   ra   Swinv2Backbone.__init__  s     #--.X]^abhbobo^pXq1rXqST#f6F6FA6M2NXq1rr*62$V__-G-GH 	 2ss   %B1c                 .    U R                   R                  $ r_   r  rk   s    r)   r  #Swinv2Backbone.get_input_embeddings  r  r(   Nr   r   r  r  rN   c           	      4   Ub  UOU R                   R                  nUb  UOU R                   R                  nUb  UOU R                   R                  nU R	                  U5      u  pgU R                  UUUSSUS9nU(       a  UR                  OUS   n	Sn
[        U R                  U	5       H  u  pXR                  ;   d  M  X4-  n
M     U(       d#  U
4nU(       a  XS   4-  nU(       a  XS   4-  nU$ [        U
U(       a  UR                  OSUR                  S9$ )	a  
Examples:

```python
>>> from transformers import AutoImageProcessor, AutoBackbone
>>> import torch
>>> from PIL import Image
>>> import httpx
>>> from io import BytesIO

>>> url = "http://images.cocodataset.org/val2017/000000039769.jpg"
>>> with httpx.stream("GET", url) as response:
...     image = Image.open(BytesIO(response.read()))

>>> processor = AutoImageProcessor.from_pretrained("microsoft/swinv2-tiny-patch4-window8-256")
>>> model = AutoBackbone.from_pretrained(
...     "microsoft/swinv2-tiny-patch4-window8-256", out_features=["stage1", "stage2", "stage3", "stage4"]
... )

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

>>> outputs = model(**inputs)
>>> feature_maps = outputs.feature_maps
>>> list(feature_maps[-1].shape)
[1, 2048, 7, 7]
```NT)r   r  r  r  r;   r   r   r8   )feature_mapsr   r   )r   r  r  r   r   r  r   rb  stage_namesout_featuresr   r   r   )rb   r   r   r  r  r  r  r   r  r   r  r  hidden_staterY   s                 r)   rg   Swinv2Backbone.forward  s0   J &1%<k$++BYBY$8$D $++JjJj 	 2C1N-TXT_T_TqTq-1__\-J*,,/!%59#  
 ;F667SU;#&t'7'7#GE)))/ $H "_F#1:-' 1:-'M%3G'//T))
 	
r(   )r   r  r  )NNN)r   r   r    r!   ra   r  r   r
   r   r   r   r   rg   r'   rp   rq   s   @r)   r  r    s    0   *.,0#'F
F
  $;F
 #Tk	F

 D[F
 
F
  ! F
r(   r  )r  r  r  r  r  )rP   F)?r"   collections.abcr   r  dataclassesr   r#   r   r    r   r  activationsr   backbone_utilsr	   r
   modeling_layersr   modeling_outputsr   modeling_utilsr   utilsr   r   r   r   utils.genericr   configuration_swinv2r   
get_loggerr   loggerr   r+   r/   r4   rH   rJ   rn   r   rZ   r   r\   rs   rv   r   r   r-  r8  rB  rL  rR  r  r  r  r  r  r  r  __all__r   r(   r)   <module>r(     s   (   !   & ! H 9 . - D D - . 
		H	% H+ H H  H H H& Hk H H* H+ H H,	U\\ e T V[VbVb  %RYY %Y-ryy Y-z(-BII (-V3 3lE>")) E>R
ryy 
bii 6  	299 	w")) wt9, 9x]
BII ]
@ PO P P< P
' P
 P
f 
e
#8 e
e
P <
#8 <
<
~ 
W
]$9 W

W
tr(   