+
    œ°³j�*  ã                   óÎ  € R t ^ RIHt ^RIHt ^RIHt ^RIHtH	t	 ]! RR7      ] ! R R]4      4       4       t
]! RR7      ] ! R	 R
]4      4       4       t]! RR7      ] ! R R]4      4       4       t]! RR7      ] ! R R]4      4       4       t]! RR7      ] ! R R]4      4       4       t]! RR7      ] ! R R]4      4       4       t]! RR7      ] ! R R]4      4       4       t. ROtR# )zSAM3 model configuration)Ústrict)ÚPreTrainedConfig)Úauto_docstring)ÚCONFIG_MAPPINGÚ
AutoConfigzfacebook/sam3)Ú
checkpointc                   ó’   a a€ ] tR t^t oRtRtRtRtRtRt	^ t
^t^tRt^tRtR	tR
t^tRtRtRtR
tRtRtV 3R ltV3R ltRtVtV ;t# )ÚSam3ViTConfiga¸  
window_size (`int`, *optional*, defaults to 24):
    Window size for windowed attention.
global_attn_indexes (`list[int]`, *optional*, defaults to `[7, 15, 23, 31]`):
    Indexes of layers with global attention.
pretrain_image_size (`int`, *optional*, defaults to 336):
    Pretrained model image size for position embedding initialization.
hidden_dropout (`float`, *optional*, defaults to 0.0):
    Dropout probability for hidden states.
Úbackbone_configÚsam3_vit_modelÚaxialé   i€  ið  Úgeluç�íµ ÷Æ°>ç        NiP  ç{®Gáz”?c                óV   <€ V P                   f
   . ROV n         \        SV `  ! R/ VB  R # )N)é   é   é   é   © )Úglobal_attn_indexesÚsuperÚ__post_init__©ÚselfÚkwargsÚ	__class__s   &,€Úl/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/sam3/configuration_sam3.pyr   ÚSam3ViTConfig.__post_init__;   s'   ø€ Ø×#Ñ#Ò+Ú'6ˆDÔ$ä‰ÒÑ' Ô'ó    c                ó  <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ S[S[ ,          ,          S[S[ S[ 3,          ,          ;R&   S[ S[S[ ,          ,          S[S[ S[ 3,          ,          ;R&   S[;R&   S[;R	&   S[S[ ,          ;R
&   S[ ;R&   S[S[ ,          R,          ;R&   S[R,          ;R&   S[ S[S[ ,          ,          S[S[ S[ 3,          ,          ;R&   S[S[ ,          ;R&   S[;R&   S[R,          ;R&   # )é   Úhidden_sizeÚintermediate_sizeÚnum_hidden_layersÚnum_attention_headsÚnum_channelsÚ
image_sizeÚ
patch_sizeÚ
hidden_actÚlayer_norm_epsÚattention_dropoutÚwindow_sizeNr   Úlayer_scale_init_valueÚpretrain_image_sizeÚhidden_dropoutÚinitializer_rangeÚrope_parameters)ÚintÚlistÚtupleÚstrÚfloatÚdict)ÚformatÚ__classdict__s   "€r   Ú__annotate__ÚSam3ViTConfig.__annotate__   s@  ø‡ ‚ ñ$ Ññ% ñ& Ñ!ñ' ñ( Ññ) ñ* Ñ!ñ+ ñ, Ññ- ñ. ‘d™3•i•¡%©©S¨¥/Õ1Ñ8ñ/ ñ0 ‘d™3•i•¡%©©S¨¥/Õ1Ñ6ñ1 ñ2 Ññ3 ñ4 Ñ ñ5 ñ6 ™s•{Ñ(ñ7 ñ8 Ññ9 ñ: ™c� TÕ)Ñ0ñ; ñ< " D�LÑ/ñ= ñ> ™t¡C�y�©5±±c°­?Õ:Ñ@ñ? ñ@ ™C•KÑ%ñA ñB Ñ#ñC ñD ˜D•[Ñ'òE r!   )r   )Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__Ú__doc__Úbase_config_keyÚ
model_typeÚdefault_rope_typer$   r%   r&   r'   r(   r)   r*   r+   r,   r-   r.   r   r/   r0   r1   r2   r3   r   Ú__annotate_func__Ú__static_attributes__Ú__classdictcell__Ú__classcell__©r   r;   s   @@r   r	   r	      s†   ù‡ € ñ	ð (€OØ!€JØÐà€KØ!ÐØÐØ!ÐØ€LØ48€JØ46€JØ€JØ €NØ%(ÐØ€KØ,0ÐØ+/ÐØ=@ÐØ"%€NØ#ÐØ#'€Oõ(÷I … r!   r	   c                   ó¢   a a€ ] tR t^Bt oRtRtRtR]/tRt	Rt
RtRtRtRtR	tV 3R
 lt]R 4       t]P&                  R 4       tV3R ltRtVtV ;t# )ÚSam3VisionConfigaÒ  
fpn_hidden_size (`int`, *optional*, defaults to 256):
    The hidden dimension of the FPN.
backbone_feature_sizes (`List[List[int]]`, *optional*, defaults to `[[288, 288], [144, 144], [72, 72]]`):
    The spatial sizes (height, width) of the feature maps from the backbone at different scales.
scale_factors (`list[float]`, *optional*, defaults to `[4.0, 2.0, 1.0, 0.5]`):
    Scale factors for FPN multi-scale features. List of scaling factors for each FPN level.
Úvision_configÚsam3_vision_modelr
   Né   r   r   r   c                óî  <€ V P                   f   . ROMV P                   V n         V P                  f   RR.^�^�.^H^H..V n        \        V P                  \        4      '       d^   V P                  P                  RR4      V P                  R&   \        V P                  R,          ,          ! R/ V P                  B V n        M%V P                  f   \        R,          ! 4       V n        \        SV `   ! R/ VB  R # )Ni   rD   r   )g      @g       @g      ð?g      à?r   )	Úscale_factorsÚbackbone_feature_sizesÚ
isinstancer
   r9   Úgetr   r   r   r   s   &,€r   r   ÚSam3VisionConfig.__post_init__\   sÒ   ø€ Ø59×5GÑ5GÒ5OÓ1ÐUY×UgÑUgˆÔØ×&Ñ&Ò.Ø,/°¨:¸¸S°zÀBÈÀ8Ð*LˆDÔ'ä�d×*Ñ*¬D×1Ò1Ø15×1EÑ1E×1IÑ1IÈ,ÐXhÓ1iˆD× Ñ  Ñ.Ü#1°$×2FÑ2FÀ|Õ2TÖ#UÑ#mÐX\×XlÑXlÑ#mˆDÕ Ø×!Ñ!Ò)Ü#1Ð2BÖ#CÓ#EˆDÔ ä‰ÒÑ' Ô'r!   c                ó.   € V P                   P                  # )z"Image size for the vision encoder.©r
   r)   ©r   s   &r   r)   ÚSam3VisionConfig.image_sizei   s   € ð ×#Ñ#×.Ñ.Ð.r!   c                ó&   € WP                   n        R# )z-Set the image size and propagate to backbone.NrW   ©r   Úvalues   &&r   r)   rY   n   s   € ð +0×ÑÖ'r!   c                ó¸   <€ V ^8„  d   Qh/ S[ S[,          R,          ;R&   S[;R&   S[R,          ;R&   S[S[,          R,          ;R&   S[;R&   S[;R&   S[;R&   # )	r#   Nr
   Úfpn_hidden_sizerR   rQ   r+   r,   r2   )r9   r   r4   r5   r8   r7   )r:   r;   s   "€r   r<   ÚSam3VisionConfig.__annotate__B   sx   ø‡ ‚ ñ$ Ñ,Õ,¨tÕ3Ñ:ñ% ñ& Ññ' ñ( ! 4�KÑ.ñ) ñ* ™•; Õ%Ñ,ñ+ ñ, Ññ- ñ. Ñ ñ/ ñ0 Ñ#ò1 r!   )r
   rR   rQ   )r>   r?   r@   rA   rB   rC   rD   r   Úsub_configsr
   r^   rR   rQ   r+   r,   r2   r   Úpropertyr)   ÚsetterrF   rG   rH   rI   rJ   s   @@r   rL   rL   B   s„   ù‡ € ñð &€OØ$€Jà˜:ð€Kð 7;€OØ€OØ*.ÐØ(,€MØ€JØ €NØ#Ðõ(ð ñ/ó ð/ð ×Ññ0ó ð0÷[ … r!   rL   c                   ó\   a € ] tR t^tt o RtRtRt^t^tRt	Rt
RtRtRt^tR	tV 3R
 ltRtV tR# )ÚSam3GeometryEncoderConfigzW
roi_size (`int`, *optional*, defaults to 7):
    ROI size for box pooling operations.
Úsam3_geometry_encoderrO   é   çš™™™™™¹?Úrelur   r   r   c                ó²   <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[S[ ,          ;R&   S[;R&   S[S[ ,          ;R&   S[;R&   S[ ;R	&   S[;R
&   # )r#   r$   Ú
num_layersr'   r%   Údropoutr+   r1   r,   Úroi_sizer2   ©r4   r8   r7   )r:   r;   s   "€r   r<   Ú&Sam3GeometryEncoderConfig.__annotate__t   s�   ø‡ ‚ ñ Ññ ñ Ññ ñ Ñ ñ ñ Ñ!ñ ñ ‘S�[Ññ ñ Ññ ñ  ™C•KÑ%ñ! ñ" Ñ ñ# ñ$ Ññ% ñ& Ñ#ò' r!   r   N)r>   r?   r@   rA   rB   rD   r$   rj   r'   r%   rk   r+   r1   r,   rl   r2   rF   rG   rH   ©r;   s   @r   rd   rd   t   sM   ø‡ € ñð
 )€Jà€KØ€JØ ÐØ!ÐØ€GØ€JØ"%€NØ €NØ€HØ#Ð÷' ƒ r!   rd   c                   óX   a € ] tR t^Št o RtRtRt^t^tRt	Rt
RtRtRtR	tV 3R
 ltRtV tR# )ÚSam3DETREncoderConfigzc
hidden_dropout (`float`, *optional*, defaults to 0.0):
    Dropout probability for hidden states.
Úsam3_detr_encoderrO   rf   rg   rh   r   r   r   c                ó¦   <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[S[ ,          ;R&   S[;R&   S[S[ ,          ;R&   S[;R&   S[;R	&   # )
r#   r$   rj   r'   r%   rk   r+   r1   r,   r2   rm   )r:   r;   s   "€r   r<   Ú"Sam3DETREncoderConfig.__annotate__Š   s‚   ø‡ ‚ ñ Ññ ñ Ññ ñ Ñ ñ ñ Ñ!ñ ñ ‘S�[Ññ ñ Ññ ñ  ™C•KÑ%ñ! ñ" Ñ ñ# ñ$ Ñ#ò% r!   r   N)r>   r?   r@   rA   rB   rD   r$   rj   r'   r%   rk   r+   r1   r,   r2   rF   rG   rH   ro   s   @r   rq   rq   Š   sH   ø‡ € ñð
 %€Jà€KØ€JØ ÐØ!ÐØ€GØ€JØ"%€NØ €NØ#Ð÷% ƒ r!   rq   c                   ó\   a € ] tR t^Ÿt o RtRtRt^t^Èt^t	Rt
RtRtRtRtR	tV 3R
 ltRtV tR# )ÚSam3DETRDecoderConfigzQ
num_queries (`int`, *optional*, defaults to 200):
    Number of object queries.
Úsam3_detr_decoderrO   rf   rg   rh   r   r   r   c                ó²   <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[S[ ,          ;R&   S[;R&   S[S[ ,          ;R&   S[;R	&   S[;R
&   # )r#   r$   rj   Únum_queriesr'   r%   rk   r+   r1   r,   r2   rm   )r:   r;   s   "€r   r<   Ú"Sam3DETRDecoderConfig.__annotate__Ÿ   s�   ø‡ ‚ ñ Ññ ñ Ññ ñ Ññ ñ Ñ ñ ñ Ñ!ñ ñ ‘S�[Ññ ñ  Ññ! ñ" ™C•KÑ%ñ# ñ$ Ñ ñ% ñ& Ñ#ò' r!   r   N)r>   r?   r@   rA   rB   rD   r$   rj   ry   r'   r%   rk   r+   r1   r,   r2   rF   rG   rH   ro   s   @r   rv   rv   Ÿ   sM   ø‡ € ñð
 %€Jà€KØ€JØ€KØ ÐØ!ÐØ€GØ€JØ"%€NØ €NØ#Ð÷' ƒ r!   rv   c                   óL   a € ] tR t^µt o RtRtRt^tRtRt	^t
RtV 3R ltRtV tR	# )
ÚSam3MaskDecoderConfigzw
num_upsampling_stages (`int`, *optional*, defaults to 3):
    Number of upsampling stages in the pixel decoder (FPN).
Úsam3_mask_decoderrO   r   r   r   c                ór   <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[;R&   S[S[ ,          ;R&   S[ ;R&   S[;R&   # )r#   r$   Únum_upsampling_stagesr,   rk   r'   r2   )r4   r8   )r:   r;   s   "€r   r<   Ú"Sam3MaskDecoderConfig.__annotate__µ   sW   ø‡ ‚ ñ Ññ ñ Ñ"ñ ñ Ñ ñ ñ ‘S�[Ññ ñ Ñ ñ ñ Ñ#ò r!   r   N)r>   r?   r@   rA   rB   rD   r$   r   r,   rk   r'   r2   rF   rG   rH   ro   s   @r   r|   r|   µ   s9   ø‡ € ñð
 %€Jà€KØ!"ÐØ €NØ€GØ ÐØ#Ð÷ ƒ r!   r|   c                   ó¶   a a€ ] tR t^Çt oRtRtRtR]R]R]	R]
R]R	]/tR
tR
tR
tR
tR
tR
tRtV 3R lt]R 4       t]P0                  R 4       tV3R ltRtVtV ;t# )Ú
Sam3Configaø  
geometry_encoder_config (`dict` or `Sam3GeometryEncoderConfig`, *optional*):
    Configuration for the geometry encoder.
detr_encoder_config (`dict` or `Sam3DETREncoderConfig`, *optional*):
    Configuration for the DETR encoder.
detr_decoder_config (`dict` or `Sam3DETRDecoderConfig`, *optional*):
    Configuration for the DETR decoder.
mask_decoder_config (`dict` or `Sam3MaskDecoderConfig`, *optional*):
    Configuration for the mask decoder.

Example:
```python
>>> from transformers import Sam3Config, Sam3Model

>>> # Initializing a SAM3 configuration
>>> configuration = Sam3Config()

>>> # Initializing a model from the configuration
>>> model = Sam3Model(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config
```
Úsam3TrM   Útext_configÚgeometry_encoder_configÚdetr_encoder_configÚdetr_decoder_configÚmask_decoder_configNr   c                ó²  <€ V P                   f   \        4       V n         \        V P                   \        4      '       d   \        R/ V P                   B V n         V P                  f)   \
        R,          ! R/ RRRRRRRR	R
^R^R^ RR/B V n        \        V P                  \        4      '       d=   \
        V P                  P                  RR4      ,          ! R/ V P                  B V n        V P                  f   \        4       V n        \        V P                  \        4      '       d   \        R/ V P                  B V n        V P                  f   \        4       V n	        \        V P                  \        4      '       d   \        R/ V P                  B V n	        V P                  f   \        4       V n        \        V P                  \        4      '       d   \        R/ V P                  B V n        V P                  f   \        4       V n        \        V P                  \        4      '       d   \        R/ V P                  B V n        \        SV `@  ! R/ VB  R # )NÚclip_text_modelÚ
vocab_sizei Á  r$   r   r%   i   Úprojection_dimi   r&   r'   Úmax_position_embeddingsr+   r   rD   r   )rM   rL   rS   r9   r„   r   rT   r…   rd   r†   rq   r‡   rv   rˆ   r|   r   r   r   s   &,€r   r   ÚSam3Config.__post_init__ö   sï  ø€ Ø×ÑÒ%Ü!1Ó!3ˆDÔÜ�d×(Ñ(¬$×/Ò/Ü!1Ñ!G°D×4FÑ4FÑ!GˆDÔà×ÑÒ#Ü-Ð.?Ö@ñ  à  %Ø! 4Ø'¨Ø$ cØ'¨Ø)¨2Ø-¨rØ  &ð	ñ ˆDÔô �d×&Ñ&¬×-Ò-Ü-¨d×.>Ñ.>×.BÑ.BÀ<ÐQbÓ.cÖdñ  Ø×"Ñ"ñ ˆDÔð ×'Ñ'Ò/Ü+DÓ+FˆDÔ(Ü�d×2Ñ2´D×9Ò9Ü+DÑ+dÀt×GcÑGcÑ+dˆDÔ(à×#Ñ#Ò+Ü'<Ó'>ˆDÔ$Ü�d×.Ñ.´×5Ò5Ü'<Ñ'X¸t×?WÑ?WÑ'XˆDÔ$à×#Ñ#Ò+Ü'<Ó'>ˆDÔ$Ü�d×.Ñ.´×5Ò5Ü'<Ñ'X¸t×?WÑ?WÑ'XˆDÔ$à×#Ñ#Ò+Ü'<Ó'>ˆDÔ$Ü�d×.Ñ.´×5Ò5Ü'<Ñ'X¸t×?WÑ?WÑ'XˆDÔ$ä‰ÒÑ' Ô'r!   c                ó.   € V P                   P                  # )zImage size for the SAM3 model.©rM   r)   rX   s   &r   r)   ÚSam3Config.image_size$  s   € ð ×!Ñ!×,Ñ,Ð,r!   c                ó&   € WP                   n        R# )z2Set the image size and propagate to vision config.Nr�   r[   s   &&r   r)   r‘   )  s   € ð ).×ÑÖ%r!   c                ó"  <€ V ^8„  d   Qh/ S[ S[,          R,          ;R&   S[ S[,          R,          ;R&   S[ S[,          R,          ;R&   S[ S[,          R,          ;R&   S[ S[,          R,          ;R&   S[ S[,          R,          ;R&   S[;R&   # )	r#   NrM   r„   r…   r†   r‡   rˆ   r2   )r9   r   r8   )r:   r;   s   "€r   r<   ÚSam3Config.__annotate__Ç   s°   ø‡ ‚ ñN Ñ*Õ*¨TÕ1Ñ8ñO ñP Ñ(Õ(¨4Õ/Ñ6ñQ ñR "Ñ$4Õ4°tÕ;ÑBñS ñT Ñ 0Õ0°4Õ7Ñ>ñU ñV Ñ 0Õ0°4Õ7Ñ>ñW ñX Ñ 0Õ0°4Õ7Ñ>ñY ñZ Ñ#ò[ r!   )r‡   r†   r…   rˆ   r„   rM   )r>   r?   r@   rA   rB   rD   Úis_compositionrL   r   rd   rq   rv   r|   r`   rM   r„   r…   r†   r‡   rˆ   r2   r   ra   r)   rb   rF   rG   rH   rI   rJ   s   @@r   r‚   r‚   Ç   s¦   ù‡ € ñð2 €JØ€NàÐ)Ø�zØ!Ð#<ØÐ4ØÐ4ØÐ4ð€Kð 59€MØ26€KØ>BÐØ:>ÐØ:>ÐØ:>ÐØ#Ðõ,(ð\ ñ-ó ð-ð ×Ññ.ó ð.÷G … r!   r‚   N)r‚   r	   rL   rd   rq   rv   r|   )rB   Úhuggingface_hub.dataclassesr   Úconfiguration_utilsr   Úutilsr   Úautor   r   r	   rL   rd   rq   rv   r|   r‚   Ú__all__r   r!   r   Ú<module>r›      sB  ðñ å .å 3Ý #ß -ñ ˜?Ô+Øô&(Ð$ó &(ó ó ,ð&(ñR ˜?Ô+Øô-0Ð'ó -0ó ó ,ð-0ñ` ˜?Ô+Øô$Ð 0ó $ó ó ,ð$ñ( ˜?Ô+Øô$Ð,ó $ó ó ,ð$ñ& ˜?Ô+Øô$Ð,ó $ó ó ,ð$ñ( ˜?Ô+Øô$Ð,ó $ó ó ,ð$ñ  ˜?Ô+Øôc.Ð!ó c.ó ó ,ðc.òL‚r!   