§
    ·¯¸j^9  ã                   ó°  — U d Z ddlmZmZmZmZmZmZmZ ddl	Z	ddl
mZ ddlmc mZ ddlmZmZ ddlmZmZ ddlmZ ddlmZ dd	lmZmZ d
gZg d¢g d¢g d¢g d¢dœZeeeeeef                  f         e d<   e G d„ dej!        ¦  «        ¦   «         Z" G d„ d
ej!        ¦  «        Z#de$deee	j%        f         fd„Z&dede'dede#fd„Z(d*dedeeef         fd„Z) e e)d¬¦  «         e)d¬¦  «         e)d¬¦  «         e)d¬¦  «         e)d¬¦  «         e)d¬¦  «         e)d¬¦  «         e)d¬¦  «        d œ¦  «        Z*ed+de'dede#fd"„¦   «         Z+ed+de'dede#fd#„¦   «         Z,ed+de'dede#fd$„¦   «         Z-ed+de'dede#fd%„¦   «         Z.ed+de'dede#fd&„¦   «         Z/ed+de'dede#fd'„¦   «         Z0ed+de'dede#fd(„¦   «         Z1ed+de'dede#fd)„¦   «         Z2dS ),z•VGG

Adapted from https://github.com/pytorch/vision 'vgg.py' (BSD-3-Clause) with a few changes for
timm functionality.

Copyright 2021 Ross Wightman
é    )ÚAnyÚDictÚListÚOptionalÚTypeÚUnionÚcastN©ÚIMAGENET_DEFAULT_MEANÚIMAGENET_DEFAULT_STD)ÚClassifierHeadÚget_device_dtypeé   )Úbuild_model_with_cfg)Úregister_notrace_module)Úregister_modelÚgenerate_default_cfgsÚVGG)é@   ÚMé€   r   é   r   r   é   r   r   r   r   r   )r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   )r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   )r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   r   )Úvgg11Úvgg13Úvgg16Úvgg19Úcfgsc                   ó¼   ‡ — e Zd ZdZdddddej        ej        ddf	ded	ed
ededede	ej
                 de	ej
                 ddfˆ fd„Zdej        dej        fd„Zˆ xZS )ÚConvMlpzxConvolutional MLP block for VGG head.

    Replaces traditional Linear layers with Conv2d layers in the classifier.
    r   é   é   ç      ð?gš™™™™™É?NÚin_featuresÚout_featuresÚkernel_sizeÚ	mlp_ratioÚ	drop_rateÚ	act_layerÚ
conv_layerÚreturnc
                 ó>  •— ||	dœ}
t          ¦   «                              ¦   «          || _        t          ||z  ¦  «        } ||||fddi|
¤Ž| _         |d¦  «        | _        t          j        |¦  «        | _         |||dfddi|
¤Ž| _	         |d¦  «        | _
        dS )a…  Initialize ConvMlp.

        Args:
            in_features: Number of input features.
            out_features: Number of output features.
            kernel_size: Kernel size for first conv layer.
            mlp_ratio: Ratio for hidden layer size.
            drop_rate: Dropout rate.
            act_layer: Activation layer type.
            conv_layer: Convolution layer type.
        ©ÚdeviceÚdtypeÚbiasTr   N)ÚsuperÚ__init__Úinput_kernel_sizeÚintÚfc1Úact1ÚnnÚDropoutÚdropÚfc2Úact2)Úselfr$   r%   r&   r'   r(   r)   r*   r.   r/   ÚddÚmid_featuresÚ	__class__s               €ú_/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/timm/models/vgg.pyr2   zConvMlp.__init__&   sº   ø€ ð. ¨Ð/Ð/ˆÝ‰Œ×ÒÑÔÐØ!,ˆÔÝ˜<¨)Ñ3Ñ4Ô4ˆØ�:˜k¨<¸ÐVÐVÈ4ÐVÐSUÐVÐVˆŒØ�I˜d‘O”OˆŒ	Ý”J˜yÑ)Ô)ˆŒ	Ø�:˜l¨L¸!ÐMÐMÀ$ÐMÈ"ÐMÐMˆŒØ�I˜d‘O”OˆŒ	ˆ	ˆ	ó    Úxc                 óÚ  — |j         d         | j        k     s|j         d         | j        k     rUt          | j        |j         d         ¦  «        t          | j        |j         d         ¦  «        f}t          j        ||¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|  	                    |¦  «        }|S )zoForward pass.

        Args:
            x: Input tensor.

        Returns:
            Output tensor.
        éþÿÿÿéÿÿÿÿ)
Úshaper3   ÚmaxÚFÚadaptive_avg_pool2dr5   r6   r9   r:   r;   )r<   rB   Úoutput_sizes      r@   ÚforwardzConvMlp.forwardG   sÁ   € ð Œ7�2Œ;˜Ô/Ò/Ð/°1´7¸2´;ÀÔAWÒ3WÐ3Wå˜tÔ5°q´w¸r´{ÑCÔCÅSÈÔI_ÐabÔahÐikÔalÑEmÔEmÐnˆKÝÔ% a¨Ñ5Ô5ˆAØ�HŠH�Q‰KŒKˆØ�IŠI�a‰LŒLˆØ�IŠI�a‰LŒLˆØ�HŠH�Q‰KŒKˆØ�IŠI�a‰LŒLˆØˆrA   )Ú__name__Ú
__module__Ú__qualname__Ú__doc__r7   ÚReLUÚConv2dr4   Úfloatr   ÚModuler2   ÚtorchÚTensorrK   Ú__classcell__©r?   s   @r@   r    r       sî   ø€ € € € € ðð ð  #Ø $Ø Ø"Ø"Ø)+¬Ø*,¬)ØØð$ð $àð$ð ð$ð ð	$ð
 ð$ð ð$ð ˜BœI”ð$ð ˜RœYœð$ð 
ð$ð $ð $ð $ð $ð $ðB˜œð ¨%¬,ð ð ð ð ð ð ð ð rA   r    c                   ó(  ‡ — e Zd ZdZddddej        ej        dddddfd	ee         d
e	de	de	de
deej                 deej                 deeej                          dede
ddfˆ fd„Zej        j        d#dedeeef         fd„¦   «         Zej        j        d$deddfd„¦   «         Zej        j        dej        fd„¦   «         Zd%d
e	dee         ddfd„Zdej        dej        fd„Zd#dej        dedej        fd „Zdej        dej        fd!„Zd&d"„Zˆ xZS )'r   z•VGG model architecture.

    Based on `Very Deep Convolutional Networks for Large-Scale Image Recognition`
    - https://arxiv.org/abs/1409.1556
    éè  é   é    r#   NÚavgg        ÚcfgÚnum_classesÚin_chansÚoutput_strider'   r)   r*   Ú
norm_layerÚglobal_poolr(   r+   c           
      ó˜  •— t          ¦   «                              ¦   «          ||dœ}|dk    sJ ‚|| _        || _        |
| _        d| _        |du| _        g | _        |}d}t          j	        }g }|D ]¯}t          |¦  «        dz
  }|dk    rD| j                             t          ||d|› �¬¦  «        ¦  «         | |d	d	¬
¦  «        gz  }|d	z  }Œ^t          t          |¦  «        } |||fdddœ|¤Ž}|�|| ||fi |¤Ž |d¬¦  «        gz  }n|| |d¬¦  «        gz  }|}Œ°t          j        |Ž | _        | j                             t          ||dt          |¦  «        dz
  › �¬¦  «        ¦  «         || _        d| _        t'          || j        df||
||dœ|¤Ž| _        t+          | j        |f|	|
dœ|¤Ž| _        |                      ¦   «          dS )a/  Initialize VGG model.

        Args:
            cfg: Configuration list defining network architecture.
            num_classes: Number of classes for classification.
            in_chans: Number of input channels.
            output_stride: Output stride of network.
            mlp_ratio: Ratio for MLP hidden layer size.
            act_layer: Activation layer type.
            conv_layer: Convolution layer type.
            norm_layer: Normalization layer type.
            global_pool: Global pooling type.
            drop_rate: Dropout rate.
        r-   r[   FNr   r   z	features.)Únum_chsÚ	reductionÚmoduleé   )r&   ÚstriderZ   )r&   ÚpaddingT)Úinplacer!   r"   )r'   r(   r)   r*   )Ú	pool_typer(   )r1   r2   r^   r_   r(   Úgrad_checkpointingÚuse_normÚfeature_infor7   Ú	MaxPool2dÚlenÚappendÚdictr	   r4   Ú
SequentialÚfeaturesÚnum_featuresÚhead_hidden_sizer    Ú
pre_logitsr   ÚheadÚ_initialize_weights)r<   r]   r^   r_   r`   r'   r)   r*   ra   rb   r(   r.   r/   r=   Úprev_chsÚ
net_strideÚ
pool_layerÚlayersÚvÚlast_idxÚconv2dr?   s                        €r@   r2   zVGG.__init__c   su  ø€ õ: 	‰Œ×ÒÑÔÐØ¨Ð/Ð/ˆØ Ò"Ð"Ð"Ð"Ø&ˆÔØ ˆŒØ"ˆŒØ"'ˆÔØ"¨$Ð.ˆŒØˆÔàˆØˆ
Ý”\ˆ
Ø"$ˆØð 	ð 	ˆAÝ˜6‘{”{ Q‘ˆHØ�CŠxˆxØÔ!×(Ò(­°hÈ*Ð]sÐiqÐ]sÐ]sÐ)tÑ)tÔ)tÑuÔuÐuØ˜:˜:°!¸AÐ>Ñ>Ô>Ð?Ñ?�Ø˜a‘�
�
å�˜a‘L”L�Ø#˜ H¨aÐP¸QÈÐPÐPÈRÐPÐP�ØÐ)Ø˜v z z°!Ð':Ð':°rÐ':Ð':¸I¸IÈdÐ<SÑ<SÔ<SÐTÑT�F�Fà˜v y y¸Ð'>Ñ'>Ô'>Ð?Ñ?�FØ��Ýœ vÐ.ˆŒØÔ× Ò ¥¨hÀ*ÐUrÕadÐekÑalÔalÐopÑapÐUrÐUrÐ!sÑ!sÔ!sÑtÔtÐtà$ˆÔØ $ˆÔÝ!ØØÔ!Øð	
ð  ØØØ!ð	
ð 	
ð ð	
ð 	
ˆŒõ #ØÔ!Øð
ð "Øð	
ð 
ð
 ð
ð 
ˆŒ	ð 	× Ò Ñ"Ô"Ð"Ð"Ð"rA   FÚcoarsec                 ó$   — t          dd¬¦  «        S )z¯Group matcher for parameter groups.

        Args:
            coarse: Whether to use coarse grouping.

        Returns:
            Dictionary of grouped parameters.
        z^features\.0z^features\.(\d+))ÚstemÚblocks)rr   )r<   r�   s     r@   Úgroup_matcherzVGG.group_matcherµ   s   € õ ˜Ð1DÐEÑEÔEÐErA   TÚenablec                 ó   — |r
J d¦   «         ‚dS )zEnable or disable gradient checkpointing.

        Args:
            enable: Whether to enable gradient checkpointing.
        z$gradient checkpointing not supportedN© )r<   r†   s     r@   Úset_grad_checkpointingzVGG.set_grad_checkpointingÂ   s   € ð ÐAÐAÐAÑAÔAˆzÐAÐArA   c                 ó   — | j         j        S )zTGet the classifier module.

        Returns:
            Classifier module.
        )rx   Úfc)r<   s    r@   Úget_classifierzVGG.get_classifierË   s   € ð ŒyŒ|ÐrA   c                 óZ   — t          | ¦  «        }|| _         | j        j        ||fi |¤Ž dS )zšReset the classifier.

        Args:
            num_classes: Number of classes for new classifier.
            global_pool: Global pooling type.
        N)r   r^   rx   Úreset)r<   r^   rb   r=   s       r@   Úreset_classifierzVGG.reset_classifierÔ   s=   € õ ˜dÑ#Ô#ˆØ&ˆÔØˆŒ	Œ˜ [Ð7Ð7°BÐ7Ð7Ð7Ð7Ð7rA   rB   c                 ó0   — |                       |¦  «        }|S )z’Forward pass through feature extraction layers.

        Args:
            x: Input tensor.

        Returns:
            Feature tensor.
        )rt   ©r<   rB   s     r@   Úforward_featureszVGG.forward_featuresß   s   € ð �MŠM˜!ÑÔˆØˆrA   rw   c                 óˆ   — |                       |¦  «        }|r|                      ||¬¦  «        n|                      |¦  «        S )zÕForward pass through head.

        Args:
            x: Input features.
            pre_logits: Return features before final linear layer.

        Returns:
            Classification logits or features.
        )rw   )rw   rx   )r<   rB   rw   s      r@   Úforward_headzVGG.forward_headë   s@   € ð �OŠO˜AÑÔˆØ6@ÐRˆt�yŠy˜ zˆyÑ2Ô2Ð2ÀdÇiÂiÐPQÁlÄlÐRrA   c                 óZ   — |                       |¦  «        }|                      |¦  «        }|S )zoForward pass.

        Args:
            x: Input tensor.

        Returns:
            Output logits.
        )r’   r”   r‘   s     r@   rK   zVGG.forwardø   s/   € ð ×!Ò! !Ñ$Ô$ˆØ×Ò˜aÑ Ô ˆØˆrA   c                 ó¨  — |                       ¦   «         D �];}t          |t          j        ¦  «        rTt          j                             |j        dd¬¦  «         |j        �%t          j                             |j        d¦  «         Œqt          |t          j	        ¦  «        rKt          j                             |j        d¦  «         t          j                             |j        d¦  «         ŒÖt          |t          j
        ¦  «        rKt          j                             |j        dd¦  «         t          j                             |j        d¦  «         �Œ=dS )zInitialize model weights.Úfan_outÚrelu)ÚmodeÚnonlinearityNr   r   g{®Gáz„?)ÚmodulesÚ
isinstancer7   rQ   ÚinitÚkaiming_normal_Úweightr0   Ú	constant_ÚBatchNorm2dÚLinearÚnormal_)r<   Úms     r@   ry   zVGG._initialize_weights  s  € à—’‘”ð 
	-ñ 
	-ˆAÝ˜!�RœYÑ'Ô'ð 	-Ý”×'Ò'¨¬°yÈvÐ'ÑVÔVÐVØ”6Ð%Ý”G×%Ò% a¤f¨aÑ0Ô0Ð0øÝ˜A�rœ~Ñ.Ô.ð -Ý”×!Ò! !¤(¨AÑ.Ô.Ð.Ý”×!Ò! !¤&¨!Ñ,Ô,Ð,Ð,Ý˜A�rœyÑ)Ô)ð -Ý”—’ ¤¨!¨TÑ2Ô2Ð2Ý”×!Ò! !¤&¨!Ñ,Ô,Ð,ùð
	-ð 
	-rA   ©F)T)N)r+   N)rL   rM   rN   rO   r7   rP   rQ   r   r   r4   rR   r   rS   r   Ústrr2   rT   ÚjitÚignoreÚboolr   r…   r‰   rŒ   r�   rU   r’   r”   rK   ry   rV   rW   s   @r@   r   r   \   s‘  ø€ € € € € ðð ð  $ØØ!#Ø"Ø)+¬Ø*,¬)Ø48Ø$Ø!ØØðP#ð P#à�c”ðP#ð ðP#ð ð	P#ð
 ðP#ð ðP#ð ˜BœI”ðP#ð ˜RœYœðP#ð !  b¤i¤Ô1ðP#ð ðP#ð ðP#ð 
ðP#ð P#ð P#ð P#ð P#ð P#ðd „YÔð
Fð 
F Dð 
F°T¸#¸s¸(´^ð 
Fð 
Fð 
Fñ Ôð
Fð „YÔðBð B¨Tð B¸Tð Bð Bð Bñ ÔðBð „YÔð ¤	ð ð ð ñ Ôðð	8ð 	8¨Cð 	8¸hÀs¼mð 	8ÐW[ð 	8ð 	8ð 	8ð 	8ð
 %¤,ð 
°5´<ð 
ð 
ð 
ð 
ðSð S˜eœlð S¸ð SÈÌð Sð Sð Sð Sð˜œð ¨%¬,ð ð ð ð ð-ð -ð -ð -ð -ð -ð -ð -rA   Ú
state_dictr+   c                 ó@  — i }|                       ¦   «         D ]†\  }}|}|                     dd¦  «        }|                     dd¦  «        }|                     dd¦  «        }d|v r|                     dd	d
d
¦  «        }d|v r|                     dddd¦  «        }|||<   Œ‡|S )z»Convert patch embedding weight from manual patchify + linear proj to conv.

    Args:
        state_dict: State dictionary to filter.

    Returns:
        Filtered state dictionary.
    zclassifier.0zpre_logits.fc1zclassifier.3zpre_logits.fc2zclassifier.6úhead.fczclassifier.0.weightrE   r   r"   zclassifier.3.weightr!   r   )ÚitemsÚreplaceÚreshape)rª   Úout_dictÚkr~   Úk_rs        r@   Ú
_filter_fnr³     s¼   € ð €HØ× Ò Ñ"Ô"ð 	ð 	‰ˆˆ1ØˆØ�kŠk˜.Ð*:Ñ;Ô;ˆØ�kŠk˜.Ð*:Ñ;Ô;ˆØ�kŠk˜.¨)Ñ4Ô4ˆØ  AÐ%Ð%Ø—	’	˜"˜c 1 aÑ(Ô(ˆAØ  AÐ%Ð%Ø—	’	˜"˜d A qÑ)Ô)ˆAØˆ�‰ˆØ€OrA   ÚvariantÚ
pretrainedÚkwargsc                 óÖ   — |                       d¦  «        d         }|                     dd¦  «        }t          t          | |ft          |         t          d|¬¦  «        t          dœ|¤Ž}|S )zÍCreate a VGG model.

    Args:
        variant: Model variant name.
        pretrained: Load pretrained weights.
        **kwargs: Additional model arguments.

    Returns:
        VGG model instance.
    Ú_r   Úout_indices)r   r   rg   rZ   é   é   T)Úflatten_sequentialr¹   )Ú	model_cfgÚfeature_cfgÚpretrained_filter_fn)ÚsplitÚpopr   r   r   rr   r³   )r´   rµ   r¶   r]   r¹   Úmodels         r@   Ú_create_vggrÃ   +  s}   € ð �-Š-˜Ñ
Ô
˜QÔ
€Cà—*’*˜]Ð,>Ñ?Ô?€KÝ ÝØØðõ �s”)Ý¨D¸kÐJÑJÔJÝ'ðð ð ðð €Eð €LrA   Ú Úurlc                 ó6   — | dddddt           t          dddd	œ|¥S )
z»Create default configuration dictionary.

    Args:
        url: Model weight URL.
        **kwargs: Additional configuration options.

    Returns:
        Configuration dictionary.
    rY   )rZ   éà   rÇ   )r"   r"   g      ì?Úbilinearz
features.0r¬   zbsd-3-clause)rÅ   r^   Ú
input_sizeÚ	pool_sizeÚcrop_pctÚinterpolationÚmeanÚstdÚ
first_convÚ
classifierÚlicenser
   )rÅ   r¶   s     r@   Ú_cfgrÒ   E  s:   € ð Ø¨=ÀvØ¨JÝ%Õ.BØ"°)Ø!ðð ð ðð rA   ztimm/)Ú	hf_hub_id)zvgg11.tv_in1kzvgg13.tv_in1kzvgg16.tv_in1kzvgg19.tv_in1kzvgg11_bn.tv_in1kzvgg13_bn.tv_in1kzvgg16_bn.tv_in1kzvgg19_bn.tv_in1kFc                 ó6   — t          di |¤Ž}t          dd| i|¤ŽS )z£VGG 11-layer model (configuration "A") from
    `"Very Deep Convolutional Networks For Large-Scale Image Recognition" <https://arxiv.org/pdf/1409.1556.pdf>`._
    r   rµ   rˆ   )r   ©rr   rÃ   ©rµ   r¶   Ú
model_argss      r@   r   r   f  ó,   € õ
 ��˜��€JÝÐDÐD¨:ÐD¸ÐDÐDÐDrA   c                 óN   — t          ddt          j        i|¤Ž}t          dd| i|¤ŽS )z·VGG 11-layer model (configuration "A") with batch normalization
    `"Very Deep Convolutional Networks For Large-Scale Image Recognition" <https://arxiv.org/pdf/1409.1556.pdf>`._
    ra   Úvgg11_bnrµ   rˆ   )rÚ   ©rr   r7   r¡   rÃ   rÖ   s      r@   rÚ   rÚ   o  ó7   € õ
 Ð:Ð:¥¤Ð:°6Ð:Ð:€JÝÐGÐG¨jÐG¸JÐGÐGÐGrA   c                 ó6   — t          di |¤Ž}t          dd| i|¤ŽS )zžVGG 13-layer model (configuration "B")
    `"Very Deep Convolutional Networks For Large-Scale Image Recognition" <https://arxiv.org/pdf/1409.1556.pdf>`._
    r   rµ   rˆ   )r   rÕ   rÖ   s      r@   r   r   x  rØ   rA   c                 óN   — t          ddt          j        i|¤Ž}t          dd| i|¤ŽS )z·VGG 13-layer model (configuration "B") with batch normalization
    `"Very Deep Convolutional Networks For Large-Scale Image Recognition" <https://arxiv.org/pdf/1409.1556.pdf>`._
    ra   Úvgg13_bnrµ   rˆ   )rß   rÛ   rÖ   s      r@   rß   rß   �  rÜ   rA   c                 ó6   — t          di |¤Ž}t          dd| i|¤ŽS )zžVGG 16-layer model (configuration "D")
    `"Very Deep Convolutional Networks For Large-Scale Image Recognition" <https://arxiv.org/pdf/1409.1556.pdf>`._
    r   rµ   rˆ   )r   rÕ   rÖ   s      r@   r   r   Š  rØ   rA   c                 óN   — t          ddt          j        i|¤Ž}t          dd| i|¤ŽS )z·VGG 16-layer model (configuration "D") with batch normalization
    `"Very Deep Convolutional Networks For Large-Scale Image Recognition" <https://arxiv.org/pdf/1409.1556.pdf>`._
    ra   Úvgg16_bnrµ   rˆ   )râ   rÛ   rÖ   s      r@   râ   râ   “  rÜ   rA   c                 ó6   — t          di |¤Ž}t          dd| i|¤ŽS )zžVGG 19-layer model (configuration "E")
    `"Very Deep Convolutional Networks For Large-Scale Image Recognition" <https://arxiv.org/pdf/1409.1556.pdf>`._
    r   rµ   rˆ   )r   rÕ   rÖ   s      r@   r   r   œ  rØ   rA   c                 óN   — t          ddt          j        i|¤Ž}t          dd| i|¤ŽS )z·VGG 19-layer model (configuration 'E') with batch normalization
    `"Very Deep Convolutional Networks For Large-Scale Image Recognition" <https://arxiv.org/pdf/1409.1556.pdf>`._
    ra   Úvgg19_bnrµ   rˆ   )rå   rÛ   rÖ   s      r@   rå   rå   ¥  rÜ   rA   )rÄ   r¥   )3rO   Útypingr   r   r   r   r   r   r	   rT   Útorch.nnr7   Útorch.nn.functionalÚ
functionalrH   Ú	timm.datar   r   Útimm.layersr   r   Ú_builderr   Ú_features_fxr   Ú	_registryr   r   Ú__all__r   r¦   r4   Ú__annotations__rS   r    r   rr   rU   r³   r©   rÃ   rÒ   Údefault_cfgsr   rÚ   r   rß   r   râ   r   rå   rˆ   rA   r@   ú<module>rò      s  ððð ð ð @Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?Ð ?à €€€Ø Ð Ð Ð Ð Ð Ø Ð Ð Ð Ð Ð Ð Ð Ð à AÐ AÐ AÐ AÐ AÐ AÐ AÐ AØ 8Ð 8Ð 8Ð 8Ð 8Ð 8Ð 8Ð 8Ø *Ð *Ð *Ð *Ð *Ð *Ø 1Ð 1Ð 1Ð 1Ð 1Ð 1Ø <Ð <Ð <Ð <Ð <Ð <Ð <Ð <àˆ'€ð NÐMÐMØVÐVÐVØeÐeÐeØtÐtÐtð	*ð *€€dˆ3��U˜3 ˜8”_Ô%Ð%Ô&ð ð ñ ð ð9ð 9ð 9ð 9ð 9ˆbŒiñ 9ô 9ñ Ôð9ðxu-ð u-ð u-ð u-ð u-ˆ"Œ)ñ u-ô u-ð u-ðp˜4ð  D¨¨e¬lÐ):Ô$;ð ð ð ð ð.˜ð ¨$ð ¸#ð À#ð ð ð ð ð4ð ˆcð  T¨#¨s¨(¤^ð ð ð ð ð* %Ð$Ø�T GÐ,Ñ,Ô,Ø�T GÐ,Ñ,Ô,Ø�T GÐ,Ñ,Ô,Ø�T GÐ,Ñ,Ô,Ø˜ wÐ/Ñ/Ô/Ø˜ wÐ/Ñ/Ô/Ø˜ wÐ/Ñ/Ô/Ø˜ wÐ/Ñ/Ô/ð	&ð 	&ñ 	ô 	€ð ðEð E�dð E¨cð E°cð Eð Eð Eñ „ðEð ðHð H˜ð H°ð H¸ð Hð Hð Hñ „ðHð ðEð E�dð E¨cð E°cð Eð Eð Eñ „ðEð ðHð H˜ð H°ð H¸ð Hð Hð Hñ „ðHð ðEð E�dð E¨cð E°cð Eð Eð Eñ „ðEð ðHð H˜ð H°ð H¸ð Hð Hð Hñ „ðHð ðEð E�dð E¨cð E°cð Eð Eð Eñ „ðEð ðHð H˜ð H°ð H¸ð Hð Hð Hñ „ðHð Hð HrA   