§
    ·¯¸j5— ã                   ó<(  — d Z ddlZddlm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mZmZmZmZmZmZmZmZm Z m!Z!m"Z" ddl#m$Z$ dd	l%m&Z& dd
l'm(Z( ddl)m*Z*m+Z+m,Z, g d¢Z-�dde.de.de.de.fd„Z/ G d„ dej0        ¦  «        Z1 G d„ dej0        ¦  «        Z2	 	 	 	 	 	 �dde.de.de.de.de.dee.         dee
ej0                          dej0        fd„Z3	 	 	 	 	 	 �dde.de.de.de.de.dee.         dee
ej0                          dej0        fd„Z4�dde5deee                  fd„Z6	 	 	 	 	 	 	 	 �dd!e	ee
e1         e
e2         f         d"f         d#e	e.d"f         d$e	e.d"f         d%e.d&e.d'e.d(e.d)e7d*e5d+e5de	ee	e8ej0        f                  eee8ef                  f         fd,„Z9 G d-„ d.ej0        ¦  «        Z:�dd/e8d0e7de:fd1„Z;�dd3e8dee8ef         fd4„Z<�dd3e8dee8ef         fd5„Z=�dd3e8dee8ef         fd6„Z>�dd3e8dee8ef         fd7„Z?�dd3e8dee8ef         fd8„Z@�dd3e8dee8ef         fd9„ZA e+i d: e>d;d<d=d>d?d@dA¬B¦  «        “dC e>d;dDd=d>d?d@dA¬B¦  «        “dE e?d;dF¬G¦  «        “dH e?d;dI¬G¦  «        “dJ e@d;dK¬G¦  «        “dL e>d;dMdA¬N¦  «        “dO e?d;dPdPdQdA¬R¦  «        “dS e?d;dT¬G¦  «        “dU e?d;dV¬G¦  «        “dW e@d;dXd?¬Y¦  «        “dZ e>d;d[¬G¦  «        “d\ e?d;dPdPdQ¬]¦  «        “d^ e>d;d_dA¬N¦  «        “d` e>d;da¬G¦  «        “db e>d;dcdA¬N¦  «        “dd e>d;dedAdfdgdhdidj¬k¦  «        “dl e?d;dm¬G¦  «        “i dn e?d;dod=d>dQd@dj¬p¦  «        “dq e?d;dr¬G¦  «        “ds e@d;dt¬G¦  «        “du e?d;dv¬G¦  «        “dw e?d;dx¬G¦  «        “dy e?d;dz¬G¦  «        “d{ e?d;d|¬G¦  «        “d} e?d;d~¬G¦  «        “d e>d;d€¬G¦  «        “d� e=d;d‚¬G¦  «        “dƒ e>d;d„¬G¦  «        “d… e>d;d†¬G¦  «        “d‡ e>d;dˆdA¬N¦  «        “d‰ e?d;dPdPd?dŠdjdA¬‹¦  «        “dŒ e?d;d�dA¬N¦  «        “dŽ e?d;d�dA¬N¦  «        “d� e@d;d‘dA¬N¦  «        “¥i d’ e>dA¬“¦  «        “d” e?d;d•¬G¦  «        “d– e?d;d—¬G¦  «        “d˜ e?d;d™¬G¦  «        “dš e@d;d›¬G¦  «        “dœ e>d;d�dAdfdgd?djdi¬ž¦  «        “dŸ e?d;d ¬G¦  «        “d¡ e?d;d¢¬G¦  «        “d£ e?d;d¤¬G¦  «        “d¥ e@d;d¦¬G¦  «        “d§ e>d;d¨dAdfdgd?djdi¬ž¦  «        “d© e>¦   «         “dª e>d;d«dAdfdgd?djdi¬ž¦  «        “d¬ e>d;d­¬G¦  «        “d® e<d;d¯d°d±¬²¦  «        “d³ e<d;d´d°d±¬²¦  «        “dµ e<d;d¶d°d±¬²¦  «        “¥i d· e<d;d¸d=d>d@d¹d°d±¬º¦  «        “d» e<d;d¼d°d±¬²¦  «        “d½ e<d;d¾d=d>d@d¹d°d±¬º¦  «        “d¿ e<d;dÀd°d±¬²¦  «        “dÁ e<d;dÂd=d>d@d¹d°d±¬º¦  «        “dÃ e<d;dÄd°d±¬²¦  «        “dÅ e<d;dÆd=d>d@d¹d°d±¬º¦  «        “dÇ e<d;dÈd°d±¬²¦  «        “dÉ e<d;dÊd=d>d@d¹d°d±¬º¦  «        “dË e>d;dÌdh¬Y¦  «        “dÍ e?d;dÎ¬G¦  «        “dÏ e?d;dÐ¬G¦  «        “dÑ e?d;dÒ¬G¦  «        “dÓ e@d;dÔ¬G¦  «        “dÕ e>d;dÖ¬G¦  «        “d× e>d;dØdA¬N¦  «        “dÙ e>¦   «         “¥i dÚ e?d;dÛ¬G¦  «        “dÜ e<d;dÝd°d±¬²¦  «        “dÞ e<d;dßd°d±¬²¦  «        “dà e<d;dád°d±¬²¦  «        “dâ e<d;dãd=d>d@d¹d°d±¬º¦  «        “dä e<d;dåd=d>d@d¹d°d±¬º¦  «        “dæ e<d;dçdèdé¬²¦  «        “dê e<d;dëdèdé¬²¦  «        “dì e<d;dídèdé¬²¦  «        “dî e<d;dïdèdé¬²¦  «        “dð e<d;dñdèdò¬²¦  «        “dó e<d;dôdèdò¬²¦  «        “dõ e<d;dödèdò¬²¦  «        “d÷ e<d;dødèdò¬²¦  «        “dù e<d;dúdèdò¬²¦  «        “dû e<d;düdèdò¬²¦  «        “dý e<d;dþdèdò¬²¦  «        “¥i dÿ e<d;�d dèdò¬²¦  «        “�d e<d;�ddèdò¬²¦  «        “�d e<d;�ddèdò¬²¦  «        “�d e<d;�ddèdò¬²¦  «        “�d e<d;�ddèdò¬²¦  «        “�d	 e>d;�d
dAdfdgd?di�¬¦  «        “�d e=d;�dd?dŠ�¬¦  «        “�d e=d;�ddAd?dŠ�¬¦  «        “�d e=d;�ddAd?dŠ�¬¦  «        “�d e=d;�ddAdfdgd?di�¬¦  «        “�d e?d;�ddA¬N¦  «        “�d e?d;�ddA¬N¦  «        “�d e@d;�ddA¬N¦  «        “�d e=d;�ddAd?dŠ�¬¦  «        “�d e=d;�ddAd?dŠ�¬¦  «        “�d  e>dAdfd?dg�¬!¦  «        “�d" e>d;�d#dAdi�d$d?dj�d%¬ž¦  «        “¥i �d& e=dA¬“¦  «        “�d' e=dA¬“¦  «        “�d( e>¦   «         “�d) e>¦   «         “�d* e?d;�d+d?¬Y¦  «        “�d, e?d;�d-d?¬Y¦  «        “�d. e@d;�d/d?¬Y¦  «        “�d0 e>d;�d1¬G¦  «        “�d2 e>dA¬“¦  «        “�d3 e>¦   «         “�d4 e>¦   «         “�d5 e>d;�d6dAdfdgd?djdi¬ž¦  «        “�d7 e>dAdfdg�¬8¦  «        “�d9 e>dAdfdg�¬8¦  «        “�d: e>d;�d;dA¬N¦  «        “�d< e>d;�d=dA¬N¦  «        “�d> e>d;�d?¬G¦  «        “¥i �d@ e>¦   «         “�dA e?d;�dB¬G¦  «        “�dC e?d;�dDdA¬N¦  «        “�dE e>d;dAd?dj�¬F¦  «        “�dG e>d;dAd?dj�¬F¦  «        “�dH e>d;d?dŠ�dIdidjdA�¬J¦  «        “�dK e>d;dAdj�¬L¦  «        “�dM e<d;�dNdA�dO�dPdj�¬Q¦  «        “�dR e<d;�dS�dNdAd?di�d$�dPdj�¬T¦	  «	        “�dU e>d;�dSdAd?dj�¬V¦  «        “�dW e>d;�dSdAd?dj�¬V¦  «        “�dX e>d;�dSdAd?dj�¬V¦  «        “�dY e>d;�dSdAd?dj�¬V¦  «        “�dZ e>¦   «         “�d[ e>d;�d\¬G¦  «        “�d] e>dA¬“¦  «        “�d^ e>dA¬“¦  «        “¥i �d_ e>dA¬“¦  «        “�d` e?d;�da¬G¦  «        “�db e>dA¬“¦  «        “�dc e?d;�dddA¬N¦  «        “�de e<d;�df�dg�dh�did@�dNdA�¬j¦  «        “�dk e<d;�dl�dmd>dhdŠ�dNdA�¬j¦  «        “�dn e<d;�dodfdgdjdi�dNdA�¬j¦  «        “�dp e<d;�dqdfdgdjdi�dNdA�¬j¦  «        “�dr e<d;�dsdfdgdj�d%�dNdA�¬j¦  «        “�dt e<d;�dudŠ�dIdj�dP�dNdA�¬j¦  «        “�dv e<d;�dwdi�d$dj�dx�dNdA�¬j¦  «        “�dy eAd;�dz¬G¦  «        “�d{ eAd;�d|¬G¦  «        “�d} eAd;�d~¬G¦  «        “�d eAd;�d€¬G¦  «        “�d� eAd;�d‚¬G¦  «        “�dƒ eAd;�d„dA¬N¦  «        “¥i �d… eAd;�d†dA¬N¦  «        “�d‡ eAd;�dˆdA¬N¦  «        “�d‰ eAd;�dŠdA¬N¦  «        “�d‹ eAd;�dŒdA¬N¦  «        “�d� eAd;�dŽdA¬N¦  «        “�d� eAd;�d�dA¬N¦  «        “�d‘ eAd;�d’dA¬N¦  «        “�d“ eAd;�d”dA¬N¦  «        “�d• eAd;�d–¬G¦  «        “�d— eAd;�d˜¬G¦  «        “�d™ eAd;�dš¬G¦  «        “�d› eAd;�dœ¬G¦  «        “�d� eAd;�dž¬G¦  «        “�dŸ eAd;�d ¬G¦  «        “�d¡ eAd;�d¢dA¬N¦  «        “�d£ e<d;dPdPd?�dg�dhdA�¬¤¦  «        “¥¦  «        ZBe*�dd0e7de:f�d¥„¦   «         ZCe*�dd0e7de:f�d¦„¦   «         ZDe*�dd0e7de:f�d§„¦   «         ZEe*�dd0e7de:f�d¨„¦   «         ZFe*�dd0e7de:f�d©„¦   «         ZGe*�dd0e7de:f�dª„¦   «         ZHe*�dd0e7de:f�d«„¦   «         ZIe*�dd0e7de:f�d¬„¦   «         ZJe*�dd0e7de:f�d­„¦   «         ZKe*�dd0e7de:f�d®„¦   «         ZLe*�dd0e7de:f�d¯„¦   «         ZMe*�dd0e7de:f�d°„¦   «         ZNe*�dd0e7de:f�d±„¦   «         ZOe*�dd0e7de:f�d²„¦   «         ZPe*�dd0e7de:f�d³„¦   «         ZQe*�dd0e7de:f�d´„¦   «         ZRe*�dd0e7de:f�dµ„¦   «         ZSe*�dd0e7de:f�d¶„¦   «         ZTe*�dd0e7de:f�d·„¦   «         ZUe*�dd0e7de:f�d¸„¦   «         ZVe*�dd0e7de:f�d¹„¦   «         ZWe*�dd0e7de:f�dº„¦   «         ZXe*�dd0e7de:f�d»„¦   «         ZYe*�dd0e7de:f�d¼„¦   «         ZZe*�dd0e7de:f�d½„¦   «         Z[e*�dd0e7de:f�d¾„¦   «         Z\e*�dd0e7de:f�d¿„¦   «         Z]e*�dd0e7de:f�dÀ„¦   «         Z^e*�dd0e7de:f�dÁ„¦   «         Z_e*�dd0e7de:f�dÂ„¦   «         Z`e*�dd0e7de:f�dÃ„¦   «         Zae*�dd0e7de:f�dÄ„¦   «         Zbe*�dd0e7de:f�dÅ„¦   «         Zce*�dd0e7de:f�dÆ„¦   «         Zde*�dd0e7de:f�dÇ„¦   «         Zee*�dd0e7de:f�dÈ„¦   «         Zfe*�dd0e7de:f�dÉ„¦   «         Zge*�dd0e7de:f�dÊ„¦   «         Zhe*�dd0e7de:f�dË„¦   «         Zie*�dd0e7de:f�dÌ„¦   «         Zje*�dd0e7de:f�dÍ„¦   «         Zke*�dd0e7de:f�dÎ„¦   «         Zle*�dd0e7de:f�dÏ„¦   «         Zme*�dd0e7de:f�dÐ„¦   «         Zne*�dd0e7de:f�dÑ„¦   «         Zoe*�dd0e7de:f�dÒ„¦   «         Zpe*�dd0e7de:f�dÓ„¦   «         Zqe*�dd0e7de:f�dÔ„¦   «         Zre*�dd0e7de:f�dÕ„¦   «         Zse*�dd0e7de:f�dÖ„¦   «         Zte*�dd0e7de:f�d×„¦   «         Zue*�dd0e7de:f�dØ„¦   «         Zve*�dd0e7de:f�dÙ„¦   «         Zwe*�dd0e7de:f�dÚ„¦   «         Zxe*�dd0e7de:f�dÛ„¦   «         Zye*�dd0e7de:f�dÜ„¦   «         Zze*�dd0e7de:f�dÝ„¦   «         Z{e*�dd0e7de:f�dÞ„¦   «         Z|e*�dd0e7de:f�dß„¦   «         Z}e*�dd0e7de:f�dà„¦   «         Z~e*�dd0e7de:f�dá„¦   «         Ze*�dd0e7de:f�dâ„¦   «         Z€e*�dd0e7de:f�dã„¦   «         Z�e*�dd0e7de:f�dä„¦   «         Z‚e*�dd0e7de:f�då„¦   «         Zƒe*�dd0e7de:f�dæ„¦   «         Z„e*�dd0e7de:f�dç„¦   «         Z…e*�dd0e7de:f�dè„¦   «         Z†e*�dd0e7de:f�dé„¦   «         Z‡e*�dd0e7de:f�dê„¦   «         Zˆe*�dd0e7de:f�dë„¦   «         Z‰e*�dd0e7de:f�dì„¦   «         ZŠe*�dd0e7f�dí„¦   «         Z‹e*�dd0e7de:f�dî„¦   «         ZŒe*�dd0e7de:f�dï„¦   «         Z�e*�dd0e7de:f�dð„¦   «         ZŽe*�dd0e7de:f�dñ„¦   «         Z�e*�dd0e7de:f�dò„¦   «         Z�e*�dd0e7de:f�dó„¦   «         Z‘e*�dd0e7de:f�dô„¦   «         Z’e*�dd0e7de:f�dõ„¦   «         Z“ e,e”i �död³“�d÷dµ“�død»“�dùd¿“�dúdÜ“�dûdæ“�düdæ“�dýdæ“�dþdæ“�dÿdð“�d dó“�ddõ“�dd÷“�ddù“�ddû“�ddý“�ddÿ“i �d�d“�d�d“�d	�d“�d
�d“�d�dy“�d�d{“�d�d}“�d�d“�d�d�“�d�dƒ“�d�d…“�d�d‡“�d�d‰“�d�d‹“�d�d�“�d�d�“�d�d‘“¥�d“�d•�d—�d™�d›�d��dŸ�d¡�d�dœ	¥¦  «         dS (   a*  PyTorch ResNet

This started as a copy of https://github.com/pytorch/vision 'resnet.py' (BSD-3-Clause) with
additional dropout and dynamic global avg/max pool.

ResNeXt, SE-ResNeXt, SENet, and MXNet Gluon stem/downsample variants, tiered stems added by Ross Wightman

Copyright 2019, Ross Wightman
é    N)Úpartial)ÚAnyÚDictÚListÚOptionalÚTupleÚTypeÚUnion©ÚIMAGENET_DEFAULT_MEANÚIMAGENET_DEFAULT_STD)ÚDropBlock2dÚDropPathÚAvgPool2dSameÚ
BlurPool2dÚ	LayerTypeÚcreate_attnÚget_attnÚget_act_layerÚget_norm_layerÚcreate_classifierÚ	create_aaÚ	to_ntupleÚget_device_dtypeé   )Úbuild_model_with_cfg)Úfeature_take_indices)Úcheckpoint_seq)Úregister_modelÚgenerate_default_cfgsÚregister_model_deprecations)ÚResNetÚ
BasicBlockÚ
BottleneckÚkernel_sizeÚstrideÚdilationÚreturnc                 ó(   — |dz
  || dz
  z  z   dz  }|S )Nr   é   © )r%   r&   r'   Úpaddings       úb/root/private/project/pallet_detection_v2/.venv/lib/python3.11/site-packages/timm/models/resnet.pyÚget_paddingr.      s#   € Ø˜‘
˜h¨+¸©/Ñ:Ñ:¸qÑ@€GØ€Nó    c            !       ó’  ‡ — e Zd ZdZdZdddddddej        ej        ddddddfdededede	ej
                 d	ed
ededede	e         deej
                 deej
                 de	eej
                          de	eej
                          de	eej
                          de	ej
                 ddf ˆ fd„Zdd„Zdej        dej        fd„Zˆ xZS )r#   zoBasic residual block for ResNet.

    This is the standard residual block used in ResNet-18 and ResNet-34.
    r   Né@   ÚinplanesÚplanesr&   Ú
downsampleÚcardinalityÚ
base_widthÚreduce_firstr'   Úfirst_dilationÚ	act_layerÚ
norm_layerÚ
attn_layerÚaa_layerÚ
drop_blockÚ	drop_pathr(   c           	      ó¤  •— ||dœ}t          ¦   «                              ¦   «          |dk    s
J d¦   «         ‚|dk    s
J d¦   «         ‚||z  }|| j        z  }|	p|}	|duo|dk    p|	|k    }t          j        ||fd|rdn||	|	d	d
œ|¤Ž| _         ||fi |¤Ž| _        |�
 |¦   «         nt          j        ¦   «         | _         |
d¬¦  «        | _	        t          |f|||dœ|¤Ž| _        t          j        ||fd||d	dœ|¤Ž| _         ||fi |¤Ž| _        t          ||fi |¤Ž| _         |
d¬¦  «        | _        || _        || _        || _        || _        dS )á¨  
        Args:
            inplanes: Input channel dimensionality.
            planes: Used to determine output channel dimensionalities.
            stride: Stride used in convolution layers.
            downsample: Optional downsample layer for residual path.
            cardinality: Number of convolution groups.
            base_width: Base width used to determine output channel dimensionality.
            reduce_first: Reduction factor for first convolution output width of residual blocks.
            dilation: Dilation rate for convolution layers.
            first_dilation: Dilation rate for first convolution layer.
            act_layer: Activation layer class.
            norm_layer: Normalization layer class.
            attn_layer: Attention layer class.
            aa_layer: Anti-aliasing layer class.
            drop_block: DropBlock layer class.
            drop_path: Optional DropPath layer instance.
        ©ÚdeviceÚdtyper   z)BasicBlock only supports cardinality of 1r1   z/BasicBlock does not support changing base widthNr*   é   F)r%   r&   r,   r'   ÚbiasT©Úinplace©Úchannelsr&   Úenable)r%   r,   r'   rE   )ÚsuperÚ__init__Ú	expansionÚnnÚConv2dÚconv1Úbn1ÚIdentityr=   Úact1r   ÚaaÚconv2Úbn2r   ÚseÚact2r4   r&   r'   r>   )Úselfr2   r3   r&   r4   r5   r6   r7   r'   r8   r9   r:   r;   r<   r=   r>   rB   rC   ÚddÚfirst_planesÚ	outplanesÚuse_aaÚ	__class__s                         €r-   rL   zBasicBlock.__init__)   sñ  ø€ ðL ¨Ð/Ð/ˆÝ‰Œ×ÒÑÔÐà˜aÒÐÐÐ!LÑÔÐØ˜RÒÐÐÐ!RÑÔÐØ Ñ-ˆØ˜Tœ^Ñ+ˆ	Ø'Ð3¨8ˆØ Ð%ÐU¨6°Qª;Ð+T¸.ÈHÒ:Tˆå”YØØð	
ð ØÐ*�1�1 FØ"Ø#Øð	
ð 	
ð ð	
ð 	
ˆŒ
ð �:˜lÐ1Ð1¨bÐ1Ð1ˆŒØ*4Ð*@˜*˜*™,œ,˜,ÅbÄkÁmÄmˆŒØ�I dÐ+Ñ+Ô+ˆŒ	Ý˜HÐ`¨|ÀFÐSYÐ`Ð`Ð]_Ð`Ð`ˆŒå”YØØð
ð ØØØð
ð 
ð ð
ð 
ˆŒ
ð �:˜iÐ.Ð.¨2Ð.Ð.ˆŒå˜j¨)Ð:Ð:°rÐ:Ð:ˆŒà�I dÐ+Ñ+Ô+ˆŒ	Ø$ˆŒØˆŒØ ˆŒØ"ˆŒˆˆr/   c                 óˆ   — t          | j        dd¦  «        �+t          j                             | j        j        ¦  «         dS dS ©zLInitialize the last batch norm layer weights to zero for better convergence.ÚweightN)ÚgetattrrV   rN   ÚinitÚzeros_ra   ©rY   s    r-   Úzero_init_lastzBasicBlock.zero_init_last{   ó<   € å�4”8˜X tÑ,Ô,Ð8ÝŒG�NŠN˜4œ8œ?Ñ+Ô+Ð+Ð+Ð+ð 9Ð8r/   Úxc                 ó  — |}|                       |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }| j        �|                      |¦  «        }| j        �|                      |¦  «        }| j	        �|  	                    |¦  «        }||z  }|  
                    |¦  «        }|S ©N)rP   rQ   r=   rS   rT   rU   rV   rW   r>   r4   rX   ©rY   rh   Úshortcuts      r-   ÚforwardzBasicBlock.forward€   sØ   € Øˆà�JŠJ�q‰MŒMˆØ�HŠH�Q‰KŒKˆØ�OŠO˜AÑÔˆØ�IŠI�a‰LŒLˆØ�GŠG�A‰JŒJˆà�JŠJ�q‰MŒMˆØ�HŠH�Q‰KŒKˆàŒ7ÐØ—’˜‘
”
ˆAàŒ>Ð%Ø—’˜qÑ!Ô!ˆAàŒ?Ð&Ø—’ xÑ0Ô0ˆHØ	ˆX‰ˆØ�IŠI�a‰LŒLˆàˆr/   ©r(   N©Ú__name__Ú
__module__Ú__qualname__Ú__doc__rM   rN   ÚReLUÚBatchNorm2dÚintr   ÚModuler	   rL   rf   ÚtorchÚTensorrm   Ú__classcell__©r^   s   @r-   r#   r#   "   sµ  ø€ € € € € ðð ð €Ið Ø.2Ø Ø Ø !ØØ,0Ø)+¬Ø*,¬.Ø48Ø26Ø48Ø-1ØØð%P#ð P#àðP#ð ðP#ð ð	P#ð
 ! ¤Ô+ðP#ð ðP#ð ðP#ð ðP#ð ðP#ð % SœMðP#ð ˜BœI”ðP#ð ˜RœYœðP#ð !  b¤i¤Ô1ðP#ð ˜t B¤IœÔ/ðP#ð !  b¤i¤Ô1ðP#ð    ¤	Ô*ð!P#ð& 
ð'P#ð P#ð P#ð P#ð P#ð P#ðd,ð ,ð ,ð ,ð
˜œð ¨%¬,ð ð ð ð ð ð ð ð r/   r#   c            !       ó’  ‡ — e Zd ZdZdZdddddddej        ej        ddddddfdededed	e	ej
                 d
edededede	e         deej
                 deej
                 de	eej
                          de	eej
                          de	eej
                          de	ej
                 ddf ˆ fd„Zdd„Zdej        dej        fd„Zˆ xZS )r$   z{Bottleneck residual block for ResNet.

    This is the bottleneck block used in ResNet-50, ResNet-101, and ResNet-152.
    é   r   Nr1   r2   r3   r&   r4   r5   r6   r7   r'   r8   r9   r:   r;   r<   r=   r>   r(   c           
      ó*  •— ||dœ}t          ¦   «                              ¦   «          t          t          j        ||dz  z  ¦  «        |z  ¦  «        }||z  }|| j        z  }|	p|}	|duo|dk    p|	|k    }t          j        ||fdddœ|¤Ž| _         ||fi |¤Ž| _	         |
d¬	¦  «        | _
        t          j        ||fd
|rdn||	|	|ddœ|¤Ž| _         ||fi |¤Ž| _        |�
 |¦   «         nt          j        ¦   «         | _         |
d¬	¦  «        | _        t!          |f|||dœ|¤Ž| _        t          j        ||fdddœ|¤Ž| _         ||fi |¤Ž| _        t)          ||fi |¤Ž| _         |
d¬	¦  «        | _        || _        || _        || _        || _        dS )r@   rA   r1   Nr*   r   F)r%   rE   TrF   rD   )r%   r&   r,   r'   ÚgroupsrE   rH   )rK   rL   rv   ÚmathÚfloorrM   rN   rO   rP   rQ   rS   rU   rV   rR   r=   rX   r   rT   Úconv3Úbn3r   rW   Úact3r4   r&   r'   r>   )rY   r2   r3   r&   r4   r5   r6   r7   r'   r8   r9   r:   r;   r<   r=   r>   rB   rC   rZ   Úwidthr[   r\   r]   r^   s                          €r-   rL   zBottleneck.__init__¡   s  ø€ ðL ¨Ð/Ð/ˆÝ‰Œ×ÒÑÔÐå•D”J˜v¨°b©Ñ9Ñ:Ô:¸[ÑHÑIÔIˆØ Ñ,ˆØ˜Tœ^Ñ+ˆ	Ø'Ð3¨8ˆØ Ð%ÐU¨6°Qª;Ð+T¸.ÈHÒ:Tˆå”Y˜x¨ÐWÀ1È5ÐWÐWÐTVÐWÐWˆŒ
Ø�:˜lÐ1Ð1¨bÐ1Ð1ˆŒØ�I dÐ+Ñ+Ô+ˆŒ	å”YØØð

ð ØÐ*�1�1 FØ"Ø#ØØð

ð 

ð ð

ð 

ˆŒ
ð �:˜eÐ*Ð* rÐ*Ð*ˆŒØ*4Ð*@˜*˜*™,œ,˜,ÅbÄkÁmÄmˆŒØ�I dÐ+Ñ+Ô+ˆŒ	Ý˜HÐY¨u¸VÈFÐYÐYÐVXÐYÐYˆŒå”Y˜u iÐQ¸QÀUÐQÐQÈbÐQÐQˆŒ
Ø�:˜iÐ.Ð.¨2Ð.Ð.ˆŒå˜j¨)Ð:Ð:°rÐ:Ð:ˆŒà�I dÐ+Ñ+Ô+ˆŒ	Ø$ˆŒØˆŒØ ˆŒØ"ˆŒˆˆr/   c                 óˆ   — t          | j        dd¦  «        �+t          j                             | j        j        ¦  «         dS dS r`   )rb   rƒ   rN   rc   rd   ra   re   s    r-   rf   zBottleneck.zero_init_lastï   rg   r/   rh   c                 óŠ  — |}|                       |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|  	                    |¦  «        }| j
        �|  
                    |¦  «        }| j        �|                      |¦  «        }| j        �|                      |¦  «        }||z  }|                      |¦  «        }|S rj   )rP   rQ   rS   rU   rV   r=   rX   rT   r‚   rƒ   rW   r>   r4   r„   rk   s      r-   rm   zBottleneck.forwardô   s  € Øˆà�JŠJ�q‰MŒMˆØ�HŠH�Q‰KŒKˆØ�IŠI�a‰LŒLˆà�JŠJ�q‰MŒMˆØ�HŠH�Q‰KŒKˆØ�OŠO˜AÑÔˆØ�IŠI�a‰LŒLˆØ�GŠG�A‰JŒJˆà�JŠJ�q‰MŒMˆØ�HŠH�Q‰KŒKˆàŒ7ÐØ—’˜‘
”
ˆAàŒ>Ð%Ø—’˜qÑ!Ô!ˆAàŒ?Ð&Ø—’ xÑ0Ô0ˆHØ	ˆX‰ˆØ�IŠI�a‰LŒLˆàˆr/   rn   ro   r{   s   @r-   r$   r$   š   sµ  ø€ € € € € ðð ð €Ið Ø.2Ø Ø Ø !ØØ,0Ø)+¬Ø*,¬.Ø48Ø26Ø48Ø-1ØØð%L#ð L#àðL#ð ðL#ð ð	L#ð
 ! ¤Ô+ðL#ð ðL#ð ðL#ð ðL#ð ðL#ð % SœMðL#ð ˜BœI”ðL#ð ˜RœYœðL#ð !  b¤i¤Ô1ðL#ð ˜t B¤IœÔ/ðL#ð !  b¤i¤Ô1ðL#ð    ¤	Ô*ð!L#ð& 
ð'L#ð L#ð L#ð L#ð L#ð L#ð\,ð ,ð ,ð ,ð
˜œð ¨%¬,ð ð ð ð ð ð ð ð r/   r$   Úin_channelsÚout_channelsr8   r:   c	           
      óÞ   — ||dœ}	|pt           j        }|dk    r|dk    rdn|}|dk    r|p|nd}t          |||¦  «        }
t          j        t          j        | ||f||
|ddœ|	¤Ž ||fi |	¤ŽgŽ S )NrA   r   F)r&   r,   r'   rE   )rN   ru   r.   Ú
SequentialrO   )rˆ   r‰   r%   r&   r'   r8   r:   rB   rC   rZ   Úps              r-   Údownsample_convr�     sÁ   € ð  UÐ	+Ð	+€BØÐ-�rœ~€JØ ’{�{ x°1¢} }�!�!¸+€KØ5@À1²_°_�nÐ0¨øÈ!€NÝ�K ¨Ñ8Ô8€AåŒ=Ý
Œ	ØØØð		
ð ØØ#Øð		
ð 		
ð ð		
ð 		
ð 	ˆ
�<Ð&Ð& 2Ð&Ð&ðð ð r/   c	           
      ó4  — ||dœ}	|pt           j        }|dk    r|nd}
|dk    r|dk    rt          j        ¦   «         }n.|
dk    r|dk    rt          nt           j        } |d|
dd¬¦  «        }t          j        |t          j        | |dfddddœ|	¤Ž ||fi |	¤ŽgŽ S )	NrA   r   r*   TF)Ú	ceil_modeÚcount_include_padr   ©r&   r,   rE   )rN   ru   rR   r   Ú	AvgPool2dr‹   rO   )rˆ   r‰   r%   r&   r'   r8   r:   rB   rC   rZ   Ú
avg_strideÚpoolÚavg_pool_fns                r-   Údownsample_avgr–   2  sÎ   € ð  UÐ	+Ð	+€BØÐ-�rœ~€JØ# qš=˜=��¨a€JØ�‚{€{�x 1’}�}ÝŒ{‰}Œ}ˆˆà'1°Q¢ ¸8Àaº<¸<•m�mÍRÌ\ˆØˆ{˜1˜j°DÈEÐRÑRÔRˆåŒ=ØÝ
Œ	�+˜|¨QÐV°qÀ!È%ÐVÐVÐSUÐVÐVØˆ
�<Ð&Ð& 2Ð&Ð&ðð ð r/   ç        Ú	drop_probc           	      óv   — dd| rt          t          | dd¬¦  «        nd| rt          t          | dd¬¦  «        ndgS )zÅCreate DropBlock layer instances for each stage.

    Args:
        drop_prob: Drop probability for DropBlock.

    Returns:
        List of DropBlock partial instances or None for each stage.
    Né   ç      Ð?)r˜   Ú
block_sizeÚgamma_scalerD   ç      ð?)r   r   )r˜   s    r-   Údrop_blocksrŸ   M  sS   € ð 	ˆdØU^Ðh�• y¸QÈDÐQÑQÔQÐQÐdhØU^Ðh�• y¸QÈDÐQÑQÔQÐQÐdhðjð jr/   é    FÚ	block_fns.rI   Úblock_repeatsr2   r7   Úoutput_strideÚdown_kernel_sizeÚavg_downÚdrop_block_rateÚdrop_path_ratec                 ón  — |
|dœ}g }g }t          |¦  «        }d}d}dx}}t          t          | ||t          |¦  «        ¦  «        ¦  «        D �]d\  }\  }}}}d|dz   › �}|dk    rdnd}||k    r||z  }d}n||z  }d}|dk    s|||j        z  k    rIt          d|||j        z  |||||                     d¦  «        d	œ|¤Ž}|rt          di |¤Žnt          di |¤Ž}t          d|||d
œ|¤Ž}g }t          |¦  «        D ]k} | dk    r|nd}| dk    r|nd}|	|z  |dz
  z  }!| 
                     |||||f||!dk    rt          |!¦  «        nddœ|¤|¤Ž¦  «         |}||j        z  }|dz  }Œl| 
                    |t          j        |Ž f¦  «         | 
                    t          |||¬¦  «        ¦  «         �Œf||fS )a  Create ResNet stages with specified block configurations.

    Args:
        block_fns: Block class to use for each stage.
        channels: Number of channels for each stage.
        block_repeats: Number of blocks to repeat for each stage.
        inplanes: Number of input channels.
        reduce_first: Reduction factor for first convolution in each stage.
        output_stride: Target output stride of network.
        down_kernel_size: Kernel size for downsample layers.
        avg_down: Use average pooling for downsample.
        drop_block_rate: DropBlock drop rate.
        drop_path_rate: Drop path rate for stochastic depth.
        **kwargs: Additional arguments passed to block constructors.

    Returns:
        Tuple of stage modules list and feature info list.
    rA   r   r}   r   Úlayerr*   Nr:   )rˆ   r‰   r%   r&   r'   r8   r:   )r7   r'   r=   r—   )r8   r>   ©Únum_chsÚ	reductionÚmoduler+   )ÚsumÚ	enumerateÚziprŸ   rM   ÚdictÚgetr–   r�   ÚrangeÚappendr   rN   r‹   )"r¡   rI   r¢   r2   r7   r£   r¤   r¥   r¦   r§   rB   rC   ÚkwargsrZ   ÚstagesÚfeature_infoÚnet_num_blocksÚnet_block_idxÚ
net_strider'   Úprev_dilationÚ	stage_idxÚblock_fnr3   Ú
num_blocksÚdbÚ
stage_namer&   r4   Údown_kwargsÚblock_kwargsÚblocksÚ	block_idxÚ	block_dprs"                                     r-   Úmake_blocksrÆ   \  sÆ  € ðB  UÐ	+Ð	+€BØ€FØ€LÝ˜Ñ'Ô'€NØ€MØ€JØ Ð €Hˆ}Ý9BÅ3ÀyÐRZÐ\iÕkvð  xGñ  lHô  lHñ  DIô  DIñ  :Jô  :Jð ,]ñ ,]Ñ5ˆ	Ñ5�H˜f j°"Ø,˜Y¨™]Ð,Ð,ˆ
Ø 1’n�n��¨!ˆØ˜Ò&Ð&Ø˜ÑˆHØˆFˆFà˜&Ñ ˆJàˆ
Ø�QŠ;ˆ;˜( f¨xÔ/AÑ&AÒAÐAÝð 	Ø$Ø# hÔ&8Ñ8Ø,ØØ!Ø,Ø!Ÿ:š: lÑ3Ô3ð	ð 	ð ð	ð 	ˆKð ;CÐf�Ð6Ð6¨+Ð6Ð6Ð6ÍÐHfÐHfÐZeÐHfÐHfˆJåÐb¨ÀÐUWÐbÐbÐ[aÐbÐbˆØˆÝ˜zÑ*Ô*ð 	ð 	ˆIØ'0°A¢~ ~˜˜¸4ˆJØ(¨Aš~˜~�V�V°1ˆFØ&¨Ñ6¸.È1Ñ:LÑMˆIØ�MŠM˜(˜(ØØØØð		ð
  -Ø1:¸R²°�( 9Ñ-Ô-Ð-ÀTð	ð 	ð ð	ð ð	ð 	ñ 	ô 	ð 	ð %ˆMØ Ô 2Ñ2ˆHØ˜QÑˆMˆMà�Š�z¥2¤=°&Ð#9Ð:Ñ;Ô;Ð;Ø×Ò�D¨¸ZÐPZÐ[Ñ[Ô[Ñ\Ô\Ð\Ñ\à�<ÐÐr/   c            2       óÈ  ‡ — e Zd ZdZddddddddd	ddd	d
ej        ej        ddddddddfdeee	f         de
edf         dededededededededededededee
edf                  deded eeej                          d!ed"ed#ed$ed%eeeef                  f.ˆ fd&„Zej        j        d?d$ed'dfd(„¦   «         Zej        j        d@d)ed'eeef         fd*„¦   «         Zej        j        d?d+ed'dfd,„¦   «         Zej        j        d@d-ed'eeej        f         fd.„¦   «         ZdAdeded'dfd/„Z	 	 	 	 	 dBd1ej        d2eeeee         f                  d3ed4ed5ed6ed'eeej                 e
ej        eej                 f         f         fd7„Z 	 	 	 dCd2eeee         f         d8ed9ed'ee         fd:„Z!d1ej        d'ej        fd;„Z"d@d1ej        d<ed'ej        fd=„Z#d1ej        d'ej        fd>„Z$ˆ xZ%S )Dr"   aª  ResNet / ResNeXt / SE-ResNeXt / SE-Net

    This class implements all variants of ResNet, ResNeXt, SE-ResNeXt, and SENet that
      * have > 1 stride in the 3x3 conv layer of bottleneck
      * have conv-bn-act ordering

    This ResNet impl supports a number of stem and downsample options based on the v1c, v1d, v1e, and v1s
    variants included in the MXNet Gluon ResNetV1b model. The C and D variants are also discussed in the
    'Bag of Tricks' paper: https://arxiv.org/pdf/1812.01187. The B variant is equivalent to torchvision default.

    ResNet variants (the same modifications can be used in SE/ResNeXt models as well):
      * normal, b - 7x7 stem, stem_width = 64, same as torchvision ResNet, NVIDIA ResNet 'v1.5', Gluon v1b
      * c - 3 layer deep 3x3 stem, stem_width = 32 (32, 32, 64)
      * d - 3 layer deep 3x3 stem, stem_width = 32 (32, 32, 64), average pool in downsample
      * e - 3 layer deep 3x3 stem, stem_width = 64 (64, 64, 128), average pool in downsample
      * s - 3 layer deep 3x3 stem, stem_width = 64 (64, 64, 128)
      * t - 3 layer deep 3x3 stem, stem width = 32 (24, 48, 64), average pool in downsample
      * tn - 3 layer deep 3x3 stem, stem width = 32 (24, 32, 64), average pool in downsample

    ResNeXt
      * normal - 7x7 stem, stem_width = 64, standard cardinality and base widths
      * same c,d, e, s variants as ResNet can be enabled

    SE-ResNeXt
      * normal - 7x7 stem, stem_width = 64
      * same c, d, e, s variants as ResNet can be enabled

    SENet-154 - 3 layer deep 3x3 stem (same as v1c-v1s), stem_width = 64, cardinality=64,
        reduction by 2 on width of first bottleneck convolution, 3x3 downsample convs after first block
    éè  rD   r    Úavgr   r1   Ú F)r1   é€   é   i   Nr—   TÚblockÚlayers.Únum_classesÚin_chansr£   Úglobal_poolr5   r6   Ú
stem_widthÚ	stem_typeÚreplace_stem_poolÚblock_reduce_firstr¤   r¥   rI   r9   r:   r<   Ú	drop_rater§   r¦   rf   Ú
block_argsc                 ó‚  •— t          ¦   «                              ¦   «          ||dœ}|pt          ¦   «         }|dv sJ ‚|| _        || _        || _        d| _        t          |¦  «        }t          |¦  «        }d|
v }|r|	dz  nd}|rµ|	|	f}d|
v r
d|	d	z  z  |	f}t          j
        t          j        ||d
         dfddddœ|¤Ž ||d
         fi |¤Ž |d¬¦  «        t          j        |d
         |d         dfddddœ|¤Ž ||d         fi |¤Ž |d¬¦  «        t          j        |d         |dfddddœ|¤ŽgŽ | _        nt          j        ||fdddddœ|¤Ž| _         ||fi |¤Ž| _         |d¬¦  «        | _        t          |dd¬¦  «        g| _        |rct          j
        t!          dt          j        ||df|rdnddddœ|¤Ž|�t#          |f|ddœ|¤Žnd ||fi |¤Ž |d¬¦  «        g¦  «        Ž | _        n}|�_t'          |t          j        ¦  «        r |d¦  «        | _        nPt          j
        t          j        ddd¬¦  «         |d|ddœ|¤ŽgŽ | _        nt          j        ddd¬¦  «        | _         t-          t/          |¦  «        ¦  «        |¦  «        }t1          ||||f|||||||||||dœ|¤|¤Ž\  }} |D ]}! | j        |!Ž  Œ| j                             | ¦  «         |d         |d         j        z  x| _        | _        t=          | j        | j        fd|i|¤Ž\  | _        | _         |  !                    |¬¦  «         dS )a	  
        Args:
            block (nn.Module): class for the residual block. Options are BasicBlock, Bottleneck.
            layers (List[int]) : number of layers in each block
            num_classes (int): number of classification classes (default 1000)
            in_chans (int): number of input (color) channels. (default 3)
            output_stride (int): output stride of the network, 32, 16, or 8. (default 32)
            global_pool (str): Global pooling type. One of 'avg', 'max', 'avgmax', 'catavgmax' (default 'avg')
            cardinality (int): number of convolution groups for 3x3 conv in Bottleneck. (default 1)
            base_width (int): bottleneck channels factor. `planes * base_width / 64 * cardinality` (default 64)
            stem_width (int): number of channels in stem convolutions (default 64)
            stem_type (str): The type of stem (default ''):
                * '', default - a single 7x7 conv with a width of stem_width
                * 'deep' - three 3x3 convolution layers of widths stem_width, stem_width, stem_width * 2
                * 'deep_tiered' - three 3x3 conv layers of widths stem_width//4 * 3, stem_width, stem_width * 2
            replace_stem_pool (bool): replace stem max-pooling layer with a 3x3 stride-2 convolution
            block_reduce_first (int): Reduction factor for first convolution output width of residual blocks,
                1 for all archs except senets, where 2 (default 1)
            down_kernel_size (int): kernel size of residual block downsample path,
                1x1 for most, 3x3 for senets (default: 1)
            avg_down (bool): use avg pooling for projection skip connection between stages/downsample (default False)
            act_layer (str, nn.Module): activation layer
            norm_layer (str, nn.Module): normalization layer
            aa_layer (nn.Module): anti-aliasing layer
            drop_rate (float): Dropout probability before classifier, for training (default 0.)
            drop_path_rate (float): Stochastic depth drop-path rate (default 0.)
            drop_block_rate (float): Drop block rate (default 0.)
            zero_init_last (bool): zero-init the last weight in residual path (usually last BN affine weight)
            block_args (dict): Extra kwargs to pass through to block module
        rA   )é   é   r    FÚdeepr*   r1   ÚtieredrD   r}   r   r   r‘   TrF   é   )r%   r&   r,   rE   rS   rª   N)rI   r&   )r%   r&   r,   )r5   r6   r£   r7   r¥   r¤   r9   r:   r<   r¦   r§   éÿÿÿÿÚ	pool_type)rf   r+   )"rK   rL   r±   rÏ   rÐ   rÖ   Úgrad_checkpointingr   r   rN   r‹   rO   rP   rQ   rS   r·   Úfilterr   ÚmaxpoolÚ
issubclassr’   Ú	MaxPool2dr   ÚlenrÆ   Ú
add_moduleÚextendrM   Únum_featuresÚhead_hidden_sizer   rÑ   ÚfcÚinit_weights)#rY   rÍ   rÎ   rÏ   rÐ   r£   rÑ   r5   r6   rÒ   rÓ   rÔ   rÕ   r¤   r¥   rI   r9   r:   r<   rÖ   r§   r¦   rf   r×   rB   rC   rZ   Ú	deep_stemr2   Ústem_chsr¡   Ústage_modulesÚstage_feature_infoÚstager^   s#                                     €r-   rL   zResNet.__init__Õ  s€  ø€ õt 	‰Œ×ÒÑÔÐØ¨Ð/Ð/ˆØÐ)¥4¡6¤6ˆ
Ø Ð+Ð+Ð+Ð+Ø&ˆÔØ ˆŒØ"ˆŒØ"'ˆÔå! )Ñ,Ô,ˆ	Ý# JÑ/Ô/ˆ
ð ˜iÐ'ˆ	Ø%.Ð6�: ‘>�>°BˆØð 	mØ" JÐ/ˆHØ˜9Ð$Ð$Ø ¨q¡Ñ1°:Ð>�ÝœÝ”	˜( H¨Q¤K°ÐZ¸1ÀaÈeÐZÐZÐWYÐZÐZØ�
˜8 Aœ;Ð-Ð-¨"Ð-Ð-Ø�	 $Ð'Ñ'Ô'Ý”	˜( 1œ+ x°¤{°AÐ]¸aÈÐQVÐ]Ð]ÐZ\Ð]Ð]Ø�
˜8 Aœ;Ð-Ð-¨"Ð-Ð-Ø�	 $Ð'Ñ'Ô'Ý”	˜( 1œ+ x°ÐZ¸1ÀaÈeÐZÐZÐWYÐZÐZð)\ð ]ˆDŒJˆJõ œ 8¨XÐlÀ1ÈQÐXYÐ`eÐlÐlÐikÐlÐlˆDŒJØ�:˜hÐ-Ð-¨"Ð-Ð-ˆŒØ�I dÐ+Ñ+Ô+ˆŒ	Ý!¨(¸aÈÐOÑOÔOÐPˆÔð ð 	PÝœ=­&°Ý”	˜( H¨aÐj¸XÐ8L¸¸È1ÐVWÐ^cÐjÐjÐgiÐjÐjØJRÐJ^•	˜(ÐF¨X¸aÐFÐFÀ2ÐFÐFÐFÐdhØ�
˜8Ð*Ð* rÐ*Ð*Ø�	 $Ð'Ñ'Ô'ð	8ñ +ô +ð ˆDŒLˆLð Ð#Ý˜h­¬Ñ5Ô5ð FØ#+ 8¨A¡;¤;�D”L�Lå#%¤=Ýœ°¸1ÀaÐHÑHÔHØ ˜ÐC¨(¸1ÐCÐCÀÐCÐCð3Eð $F�D”L�Lõ  "œ|¸À!ÈQÐOÑOÔO�”ð -•I�c (™mœmÑ,Ô,¨UÑ3Ô3ˆ	Ý,7ØØØØð	-
ð
 $Ø!Ø'Ø+ØØ-ØØ!ØØ+Ø)ð-
ð -
ð  ð!-
ð" ð#-
ð -
Ñ)ˆÐ)ð& #ð 	$ð 	$ˆEØˆDŒO˜UÐ#Ð#Ð#ØÔ× Ò Ð!3Ñ4Ô4Ð4ð 5=¸R´LÀ9ÈRÄ=ÔCZÑ4ZÐZˆÔ˜DÔ1Ý$5°dÔ6GÈÔIYÐ$wÐ$wÐepÐ$wÐtvÐ$wÐ$wÑ!ˆÔ˜$œ'à×Ò¨ÐÑ8Ô8Ð8Ð8Ð8r/   r(   c                 ó:  — |                       ¦   «         D ]F\  }}t          |t          j        ¦  «        r't          j                             |j        dd¬¦  «         ŒG|r;|                      ¦   «         D ](}t          |d¦  «        r| 	                    ¦   «          Œ'dS dS )z‚Initialize model weights.

        Args:
            zero_init_last: Zero-initialize the last BN in each residual branch.
        Úfan_outÚrelu)ÚmodeÚnonlinearityrf   N)
Únamed_modulesÚ
isinstancerN   rO   rc   Úkaiming_normal_ra   ÚmodulesÚhasattrrf   )rY   rf   ÚnÚms       r-   rë   zResNet.init_weightsb  s³   € ð ×&Ò&Ñ(Ô(ð 	Wð 	W‰DˆAˆqÝ˜!�RœYÑ'Ô'ð WÝ”×'Ò'¨¬°yÈvÐ'ÑVÔVÐVøØð 	'Ø—\’\‘^”^ð 'ð '�Ý˜1Ð.Ñ/Ô/ð 'Ø×$Ò$Ñ&Ô&Ð&øð	'ð 	'ð'ð 'r/   Úcoarsec                 ó0   — t          d|rdnd¬¦  «        }|S )zâCreate regex patterns for parameter grouping.

        Args:
            coarse: Use coarse (stage-level) or fine (block-level) grouping.

        Returns:
            Dictionary mapping group names to regex patterns.
        z^conv1|bn1|maxpoolz^layer(\d+)z^layer(\d+)\.(\d+))ÚstemrÃ   )r±   )rY   rý   Úmatchers      r-   Úgroup_matcherzResNet.group_matcherq  s&   € õ Ð1ÈFÐ:m¸.¸.ÐXmÐnÑnÔnˆØˆr/   rJ   c                 ó   — || _         dS )zEnable or disable gradient checkpointing.

        Args:
            enable: Whether to enable gradient checkpointing.
        N)rà   )rY   rJ   s     r-   Úset_grad_checkpointingzResNet.set_grad_checkpointing~  s   € ð #)ˆÔÐÐr/   Ú	name_onlyc                 ó   — |rdn| j         S )z³Get the classifier module.

        Args:
            name_only: Return classifier module name instead of module.

        Returns:
            Classifier module or name.
        rê   )rê   )rY   r  s     r-   Úget_classifierzResNet.get_classifier‡  s   € ð !Ð-ˆtˆt d¤gÐ-r/   c                 óü   — t          | ¦  «        }|| _        t          | j        | j        fd|i|¤Ž\  | _        | _        | j                             | j        ¦  «         | j                             | j        ¦  «         dS )zŸReset the classifier head.

        Args:
            num_classes: Number of classes for new classifier.
            global_pool: Global pooling type.
        rß   N)r   rÏ   r   rè   rÑ   rê   ÚtrainÚtraining)rY   rÏ   rÑ   rZ   s       r-   Úreset_classifierzResNet.reset_classifier“  s|   € õ ˜dÑ#Ô#ˆØ&ˆÔÝ$5°dÔ6GÈÔIYÐ$wÐ$wÐepÐ$wÐtvÐ$wÐ$wÑ!ˆÔ˜$œ'ØÔ×Ò˜tœ}Ñ-Ô-Ð-ØŒ�Š�d”mÑ$Ô$Ð$Ð$Ð$r/   ÚNCHWrh   ÚindicesÚnormÚ
stop_earlyÚ
output_fmtÚintermediates_onlyc                 óÊ  — |dv s
J d¦   «         ‚g }t          d|¦  «        \  }}	d}
|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }|
|v r|                     |¦  «         |                      |¦  «        }d}|r
|d|	…         }|D ]9}|
dz  }
 t          | |¦  «        |¦  «        }|
|v r|                     |¦  «         Œ:|r|S ||fS )aJ  Forward features that returns intermediates.

        Args:
            x: Input image tensor.
            indices: Take last n blocks if int, all if None, select matching indices if sequence.
            norm: Apply norm layer to compatible intermediates.
            stop_early: Stop iterating over blocks when last desired intermediate hit.
            output_fmt: Shape of intermediate feature outputs.
            intermediates_only: Only return intermediate features.

        Returns:
            Features and list of intermediate features or just intermediate features.
        )r  zOutput shape must be NCHW.rš   r   ©Úlayer1Úlayer2Úlayer3Úlayer4Nr   )r   rP   rQ   rS   r´   râ   rb   )rY   rh   r  r  r  r  r  ÚintermediatesÚtake_indicesÚ	max_indexÚfeat_idxÚlayer_namesrû   s                r-   Úforward_intermediateszResNet.forward_intermediates   s  € ð, ˜YÐ&Ð&Ð&Ð(DÑ&Ô&Ð&ØˆÝ"6°q¸'Ñ"BÔ"BÑˆ�ið ˆØ�JŠJ�q‰MŒMˆØ�HŠH�Q‰KŒKˆØ�IŠI�a‰LŒLˆØ�|Ð#Ð#Ø× Ò  Ñ#Ô#Ð#Ø�LŠL˜‰OŒOˆà>ˆØð 	2Ø% j y jÔ1ˆKØð 	(ð 	(ˆAØ˜‰MˆHØ •˜˜aÑ Ô  Ñ#Ô#ˆAØ˜<Ð'Ð'Ø×$Ò$ QÑ'Ô'Ð'øàð 	!Ø Ð à�-ÐÐr/   Ú
prune_normÚ
prune_headc                 óÂ   — t          d|¦  «        \  }}d}||d…         }|D ]$}t          | |t          j        ¦   «         ¦  «         Œ%|r|                      dd¦  «         |S )aF  Prune layers not required for specified intermediates.

        Args:
            indices: Indices of intermediate layers to keep.
            prune_norm: Whether to prune normalization layers.
            prune_head: Whether to prune the classifier head.

        Returns:
            List of indices that were kept.
        rš   r  Nr   rÊ   )r   ÚsetattrrN   rR   r
  )rY   r  r  r  r  r  r  rû   s           r-   Úprune_intermediate_layersz ResNet.prune_intermediate_layersÑ  sy   € õ  #7°q¸'Ñ"BÔ"BÑˆ�iØ>ˆØ! ) * *Ô-ˆØð 	,ð 	,ˆAÝ�D˜!�Rœ[™]œ]Ñ+Ô+Ð+Ð+Øð 	)Ø×!Ò! ! RÑ(Ô(Ð(ØÐr/   c                 óö  — |                       |¦  «        }|                      |¦  «        }|                      |¦  «        }|                      |¦  «        }| j        rIt
          j                             ¦   «         s+t          | j	        | j
        | j        | j        g|d¬¦  «        }nT|  	                    |¦  «        }|  
                    |¦  «        }|                      |¦  «        }|                      |¦  «        }|S )z/Forward pass through feature extraction layers.T)Úflatten)rP   rQ   rS   râ   rà   rx   ÚjitÚis_scriptingr   r  r  r  r  ©rY   rh   s     r-   Úforward_featureszResNet.forward_featuresê  sÈ   € à�JŠJ�q‰MŒMˆØ�HŠH�Q‰KŒKˆØ�IŠI�a‰LŒLˆØ�LŠL˜‰OŒOˆàÔ"ð 	­5¬9×+AÒ+AÑ+CÔ+Cð 	Ý ¤¨T¬[¸$¼+ÀtÄ{ÐSÐUVÐ`dÐeÑeÔeˆAˆAà—’˜A‘”ˆAØ—’˜A‘”ˆAØ—’˜A‘”ˆAØ—’˜A‘”ˆAØˆr/   Ú
pre_logitsc                 óÈ   — |                       |¦  «        }| j        r.t          j        |t	          | j        ¦  «        | j        ¬¦  «        }|r|n|                      |¦  «        S )zÐForward pass through classifier head.

        Args:
            x: Feature tensor.
            pre_logits: Return features before final classifier layer.

        Returns:
            Output tensor.
        )rŒ   r	  )rÑ   rÖ   ÚFÚdropoutÚfloatr	  rê   )rY   rh   r(  s      r-   Úforward_headzResNet.forward_headú  s]   € ð ×Ò˜QÑÔˆØŒ>ð 	NÝ”	˜!�u T¤^Ñ4Ô4¸t¼}ÐMÑMÔMˆAØÐ.ˆqˆq D§G¢G¨A¡J¤JÐ.r/   c                 óZ   — |                       |¦  «        }|                      |¦  «        }|S )zForward pass.)r'  r-  r&  s     r-   rm   zResNet.forward	  s-   € à×!Ò! !Ñ$Ô$ˆØ×Ò˜aÑ Ô ˆØˆr/   )T©F)rÉ   )NFFr  F)r   FT)&rp   rq   rr   rs   rN   rt   ru   r
   r#   r$   r   rv   ÚstrÚboolr   r   r	   rw   r,  r   r   rL   rx   r$  Úignorerë   r  r  r  r
  ry   r   r  r!  r'  r-  rm   rz   r{   s   @r-   r"   r"   µ  sw  ø€ € € € € ðð ðF  $ØØ!#Ø$Ø Ø Ø ØØ&+Ø&'Ø$%Ø"Ø2EØ#%¤7Ø$&¤NØ26Ø"Ø$&Ø%'Ø#'Ø37ØØð5K9ð K9à˜ ZÐ/Ô0ðK9ð ˜#˜s˜(”OðK9ð ð	K9ð
 ðK9ð ðK9ð ðK9ð ðK9ð ðK9ð ðK9ð ðK9ð  $ðK9ð !$ðK9ð "ðK9ð ðK9ð  ˜u S¨# XœÔ/ð!K9ð" !ð#K9ð$ "ð%K9ð& ˜t B¤IœÔ/ð'K9ð( ð)K9ð* "ð+K9ð, #ð-K9ð. !ð/K9ð0 !  c¨3 h¤Ô0ð1K9ð K9ð K9ð K9ð K9ð K9ðZ „YÔð'ð '¨4ð '¸4ð 'ð 'ð 'ñ Ôð'ð „YÔð
ð 
 Dð 
°T¸#¸s¸(´^ð 
ð 
ð 
ñ Ôð
ð „YÔð)ð )¨Tð )¸Tð )ð )ð )ñ Ôð)ð „YÔð	.ð 	.¨ð 	.¸¸sÀBÄI¸~Ô9Nð 	.ð 	.ð 	.ñ Ôð	.ð%ð %¨Cð %¸cð %Èdð %ð %ð %ð %ð  8<ØØ$Ø$Ø',ð/ ð / àŒ|ð/ ð ˜e C¨¨c¬ NÔ3Ô4ð/ ð ð	/ ð
 ð/ ð ð/ ð !%ð/ ð 
ˆt�E”LÔ! 5¨¬°t¸E¼LÔ7IÐ)IÔ#JÐJÔ	Kð/ ð / ð / ð / ðf ./Ø$Ø#ð	ð à˜3  S¤	˜>Ô*ðð ðð ð	ð
 
ˆcŒðð ð ð ð2 %¤,ð °5´<ð ð ð ð ð /ð /˜eœlð /¸ð /ÈÌð /ð /ð /ð /ð˜œð ¨%¬,ð ð ð ð ð ð ð ð r/   r"   ÚvariantÚ
pretrainedc                 ó*   — t          t          | |fi |¤ŽS )zÓCreate a ResNet model.

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

    Returns:
        ResNet model instance.
    )r   r"   )r3  r4  rµ   s      r-   Ú_create_resnetr6    s   € õ  ¥¨°ÐFÐF¸vÐFÐFÐFr/   rÊ   Úurlc                 ó6   — | dddddt           t          dddd	œ|¥S )
z1Create a default configuration for ResNet models.rÈ   ©rD   éà   r:  )rÝ   rÝ   g      ì?ÚbilinearrP   rê   z
apache-2.0)r7  rÏ   Ú
input_sizeÚ	pool_sizeÚcrop_pctÚinterpolationÚmeanÚstdÚ
first_convÚ
classifierÚlicenser   ©r7  rµ   s     r-   Ú_cfgrF    s:   € ð Ø¨=ÀvØ¨JÝ%Õ.BØ¨TØðð ð ðð r/   c           	      ó8   — t          dd| it          ddifi |¤Ž¤ŽS )z2Create a configuration with bicubic interpolation.r7  r?  Úbicubicr+   ©rF  r±   rE  s     r-   Ú_tcfgrJ  +  s0   € åÐHÐH�CÐH�4 °)Ð <ÐGÐGÀÐGÐGÐHÐHÐHr/   c                 ó>   — t          dd| it          dddddœfi |¤Ž¤ŽS )z4Create a configuration for models trained with timm.r7  rH  ©rD   é   rM  çffffffî?ú3https://github.com/huggingface/pytorch-image-models)r?  Útest_input_sizeÚtest_crop_pctÚ
origin_urlr+   rI  rE  s     r-   Ú_ttcfgrS  0  sU   € åð ð �Cð �4Ø"°}ÐW[ØKð!ð !ð ð ð 
ðð ð ð ð r/   c                 óB   — t          d	d| it          dddddddœfi |¤Ž¤ŽS )
z,Create a configuration for ResNet-RS models.r7  rH  rN  rL  rž   rO  úarXiv:2110.00476)r?  r>  rP  rQ  rR  Ú	paper_idsr+   rI  rE  s     r-   Ú_rcfgrW  8  sZ   € åð ð �Cð �4Ø"°ÈÐilØKÐZlð!ð !ð ð ð 
ðð ð ð ð r/   c                 óF   — t          d
d| it          ddddddddd	œfi |¤Ž¤ŽS )z?Create a configuration for ResNet-RS models with 160x160 input.r7  rH  ©rD   é    rZ  ©rš   rš   rN  r9  rO  rU  )r?  r<  r=  r>  rP  rQ  rR  rV  r+   rI  rE  s     r-   Ú_r3cfgr\  @  s^   € åð ð �Cð �4Ø"°-ÈfØ¨]ÈTØKÐZlð!ð !ð ð ð 
ð	ð ð ð ð r/   c           	      ó:   — t          dd| it          dddœfi |¤Ž¤ŽS )z3Create a configuration for Gluon pretrained models.r7  rH  z1https://cv.gluon.ai/model_zoo/classification.html)r?  rR  r+   rI  rE  s     r-   Ú_gcfgr^  I  sP   € åð ð �Cð �4Ø"ØIð!ð !ð ð ð 
ðð ð ð ð r/   zresnet10t.c3_in1kztimm/zrhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet10t_176_c3-f3215ab1.pth)rD   é°   r_  )é   r`  rN  r9  zconv1.0)Ú	hf_hub_idr7  r<  r=  rQ  rP  rB  zresnet14t.c3_in1kzrhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet14t_176_c3-c4ed2c37.pthzresnet18.a1_in1kzqhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet18_a1_0-d63eafa0.pth)ra  r7  zresnet18.a2_in1kzqhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet18_a2_0-b61bd467.pthzresnet18.a3_in1kzqhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet18_a3_0-40c531c8.pthzresnet18d.ra2_in1kzkhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet18d_ra2-48a79e06.pth)ra  r7  rB  zresnet18d.ra4_e3600_r224_in1k)ç      à?rb  rb  gÍÌÌÌÌÌì?)ra  r@  rA  r>  rB  zresnet34.a1_in1kzqhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet34_a1_0-46f8f793.pthzresnet34.a2_in1kzqhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet34_a2_0-82d47d71.pthzresnet34.a3_in1kzqhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet34_a3_0-a20cabb6.pth)ra  r7  r>  zresnet34.bt_in1kzfhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34-43635321.pthzresnet34.ra4_e3600_r224_in1k)ra  r@  rA  r>  zresnet34d.ra2_in1kzkhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet34d_ra2-f8dcfcaf.pthzresnet26.bt_in1kzfhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26-9aa10e23.pthzresnet26d.bt_in1kzghttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet26d-69e92c46.pthzresnet26t.ra2_in1kzthttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-attn-weights/resnet26t_256_ra2-6f6fa748.pth)rD   rÌ   rÌ   )rÙ   rÙ   g®Gázî?)rD   é@  rc  rž   )ra  r7  rB  r<  r=  r>  rP  rQ  zresnet50.a1_in1kzohttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_a1_0-14fe96d1.pthzresnet50.a1h_in1kzuhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_a1h2_176-001a1197.pth)ra  r7  r<  r=  r>  rP  rQ  zresnet50.a2_in1kzqhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_a2_0-a2746f79.pthzresnet50.a3_in1kzqhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_a3_0-59cae1ef.pthzresnet50.b1k_in1kzphttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_b1k-532a802a.pthzresnet50.b2k_in1kzphttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_b2k-1ba180c1.pthzresnet50.c1_in1kzohttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_c1-5ba5e060.pthzresnet50.c2_in1kzohttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_c2-d01e05b2.pthzresnet50.d_in1kznhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_d-f39db8af.pthzresnet50.ram_in1kzlhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-weights/resnet50_ram-a26f946b.pthzresnet50.am_in1kzkhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-weights/resnet50_am-6c502b37.pthzresnet50.ra_in1kzkhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-weights/resnet50_ra-85ebb6e5.pthzresnet50.bt_in1kzkhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-weights/rw_resnet50-86acaeed.pthzresnet50d.ra2_in1kzkhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet50d_ra2-464e36ba.pthzresnet50d.ra4_e3600_r224_in1krL  )ra  r@  rA  r>  rP  rQ  rB  zresnet50d.a1_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50d_a1_0-e20cff14.pthzresnet50d.a2_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50d_a2_0-a3adc64d.pthzresnet50d.a3_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50d_a3_0-403fdfad.pthzresnet50t.untrained)rB  zresnet101.a1h_in1kzohttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet101_a1h-36d3f2aa.pthzresnet101.a1_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet101_a1_0-cdcb52a9.pthzresnet101.a2_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet101_a2_0-6edb36c7.pthzresnet101.a3_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet101_a3_0-1db14157.pthzresnet101d.ra2_in1kzlhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet101d_ra2-2803ffab.pth)ra  r7  rB  r<  r=  r>  rQ  rP  zresnet152.a1h_in1kzohttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet152_a1h-dc400468.pthzresnet152.a1_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet152_a1_0-2eee8a7a.pthzresnet152.a2_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet152_a2_0-b4c6978f.pthzresnet152.a3_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet152_a3_0-134d4688.pthzresnet152d.ra2_in1kzlhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet152d_ra2-5cac0439.pthzresnet200.untrainedzresnet200d.ra2_in1kzlhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnet200d_ra2-bdba9bf9.pthzwide_resnet50_2.racm_in1kzphttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/wide_resnet50_racm-8234f177.pthzresnet18.tv_in1kz9https://download.pytorch.org/models/resnet18-f37072fd.pthzbsd-3-clausez!https://github.com/pytorch/vision)ra  r7  rD  rR  zresnet34.tv_in1kz9https://download.pytorch.org/models/resnet34-b627a593.pthzresnet50.tv_in1kz9https://download.pytorch.org/models/resnet50-0676ba61.pthzresnet50.tv2_in1kz9https://download.pytorch.org/models/resnet50-11ad3fa6.pthgáz®Gáî?)ra  r7  r<  r=  rP  rQ  rD  rR  zresnet101.tv_in1kz:https://download.pytorch.org/models/resnet101-63fe2227.pthzresnet101.tv2_in1kz:https://download.pytorch.org/models/resnet101-cd907fc2.pthzresnet152.tv_in1kz:https://download.pytorch.org/models/resnet152-394f9c45.pthzresnet152.tv2_in1kz:https://download.pytorch.org/models/resnet152-f82ba261.pthzwide_resnet50_2.tv_in1kz@https://download.pytorch.org/models/wide_resnet50_2-95faca4d.pthzwide_resnet50_2.tv2_in1kz@https://download.pytorch.org/models/wide_resnet50_2-9ba9bcbe.pthzwide_resnet101_2.tv_in1kzAhttps://download.pytorch.org/models/wide_resnet101_2-32ee1156.pthzwide_resnet101_2.tv2_in1kzAhttps://download.pytorch.org/models/wide_resnet101_2-d733dc28.pthzresnet50_gn.a1h_in1kzrhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rsb-weights/resnet50_gn_a1h2-8fe6c4d0.pthzresnext50_32x4d.a1h_in1kzuhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rsb-weights/resnext50_32x4d_a1h-0146ab0a.pthzresnext50_32x4d.a1_in1kzxhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnext50_32x4d_a1_0-b5a91a1d.pthzresnext50_32x4d.a2_in1kzxhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnext50_32x4d_a2_0-efc76add.pthzresnext50_32x4d.a3_in1kzxhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/resnext50_32x4d_a3_0-3e450271.pthzresnext50_32x4d.ra_in1kzrhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-weights/resnext50_32x4d_ra-d733960d.pthzresnext50d_32x4d.bt_in1kznhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnext50d_32x4d-103e99f8.pthzresnext101_32x4d.untrainedzresnext101_64x4d.c1_in1kzthttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tpu-weights/resnext101_64x4d_c-0d0e0cc0.pthzresnext50_32x4d.tv_in1kz@https://download.pytorch.org/models/resnext50_32x4d-7cdf4587.pthzresnext101_32x8d.tv_in1kzAhttps://download.pytorch.org/models/resnext101_32x8d-8ba56ff5.pthzresnext101_64x4d.tv_in1kzAhttps://download.pytorch.org/models/resnext101_64x4d-173b62eb.pthzresnext50_32x4d.tv2_in1kz@https://download.pytorch.org/models/resnext50_32x4d-1a0047aa.pthzresnext101_32x8d.tv2_in1kzAhttps://download.pytorch.org/models/resnext101_32x8d-110c445d.pthz$resnext101_32x8d.fb_wsl_ig1b_ft_in1kzChttps://download.pytorch.org/models/ig_resnext101_32x8-c38310e5.pthzcc-by-nc-4.0z.https://github.com/facebookresearch/WSL-Imagesz%resnext101_32x16d.fb_wsl_ig1b_ft_in1kzDhttps://download.pytorch.org/models/ig_resnext101_32x16-c6f796b0.pthz%resnext101_32x32d.fb_wsl_ig1b_ft_in1kzDhttps://download.pytorch.org/models/ig_resnext101_32x32-e4b90b00.pthz%resnext101_32x48d.fb_wsl_ig1b_ft_in1kzDhttps://download.pytorch.org/models/ig_resnext101_32x48-3e41cc8a.pthz resnet18.fb_ssl_yfcc100m_ft_in1kzdhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnet18-d92f0530.pthzEhttps://github.com/facebookresearch/semi-supervised-ImageNet1K-modelsz resnet50.fb_ssl_yfcc100m_ft_in1kzdhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnet50-08389792.pthz'resnext50_32x4d.fb_ssl_yfcc100m_ft_in1kzjhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnext50_32x4-ddb3e555.pthz(resnext101_32x4d.fb_ssl_yfcc100m_ft_in1kzkhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnext101_32x4-dc43570a.pthz(resnext101_32x8d.fb_ssl_yfcc100m_ft_in1kzkhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnext101_32x8-2cfe2f8b.pthz)resnext101_32x16d.fb_ssl_yfcc100m_ft_in1kzlhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_supervised_resnext101_32x16-15fffa57.pthzresnet18.fb_swsl_ig1b_ft_in1kzkhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnet18-118f1556.pthzresnet50.fb_swsl_ig1b_ft_in1kzkhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnet50-16a12f1b.pthz$resnext50_32x4d.fb_swsl_ig1b_ft_in1kzqhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext50_32x4-72679e44.pthz%resnext101_32x4d.fb_swsl_ig1b_ft_in1kzrhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext101_32x4-3f87e46b.pthz%resnext101_32x8d.fb_swsl_ig1b_ft_in1kzrhttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext101_32x8-b4712904.pthz&resnext101_32x16d.fb_swsl_ig1b_ft_in1kzshttps://dl.fbaipublicfiles.com/semiweaksupervision/model_files/semi_weakly_supervised_resnext101_32x16-f3559a9c.pthzecaresnet26t.ra2_in1kznhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/ecaresnet26t_ra2-46609757.pth)ra  r7  rB  r<  r=  rQ  rP  zecaresnetlight.miil_in1kzlhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tresnet/ecaresnetlight-75a9c627.pth)ra  r7  rQ  rP  zecaresnet50d.miil_in1kzjhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tresnet/ecaresnet50d-93c81e3b.pth)ra  r7  rB  rQ  rP  zecaresnet50d_pruned.miil_in1kzlhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tresnet/ecaresnet50d_p-e4fa23c2.pthzecaresnet50t.ra2_in1kznhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/ecaresnet50t_ra2-f7ac63c4.pthzecaresnet50t.a1_in1kzuhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/ecaresnet50t_a1_0-99bd76a8.pthzecaresnet50t.a2_in1kzuhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/ecaresnet50t_a2_0-b1c7b745.pthzecaresnet50t.a3_in1kzuhttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/ecaresnet50t_a3_0-8cc311f1.pthzecaresnet101d.miil_in1kzkhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tresnet/ecaresnet101d-153dad65.pthzecaresnet101d_pruned.miil_in1kzmhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tresnet/ecaresnet101d_p-9e74cb91.pthzecaresnet200d.untrained)rB  r<  r>  r=  zecaresnet269d.ra2_in1kzshttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/ecaresnet269d_320_ra2-7baa55cb.pth)é
   rd  )rD   é`  re  zecaresnext26t_32x4d.untrainedzecaresnext50t_32x4d.untrainedzseresnet18.untrainedzseresnet34.untrainedzseresnet50.a1_in1kzshttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/seresnet50_a1_0-ffa00869.pthzseresnet50.a2_in1kzshttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/seresnet50_a2_0-850de0d9.pthzseresnet50.a3_in1kzshttps://github.com/huggingface/pytorch-image-models/releases/download/v0.1-rsb-weights/seresnet50_a3_0-317ecd56.pthzseresnet50.ra2_in1kzohttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet50_ra_224-8efdb4bb.pthzseresnet50t.untrainedzseresnet101.untrainedzseresnet152.untrainedzseresnet152d.ra2_in1kznhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnet152d_ra2-04464dd2.pthzseresnet200d.untrained)rB  r<  r=  zseresnet269d.untrainedzseresnext26d_32x4d.bt_in1kzphttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26d_32x4d-80fa48a3.pthzseresnext26t_32x4d.bt_in1kzqhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext26tn_32x4d-569cb627.pthzseresnext50_32x4d.racm_in1kzthttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/seresnext50_32x4d_racm-a304a460.pthzseresnext101_32x4d.untrainedzseresnext101_32x8d.ah_in1kzwhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tpu-weights/seresnext101_32x8d_ah-e6bc4c0a.pthzseresnext101d_32x8d.ah_in1kzxhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tpu-weights/seresnext101d_32x8d_ah-191d7b94.pthzresnetaa50d.sw_in12k_ft_in1k)ra  rB  r>  rQ  zresnetaa101d.sw_in12k_ft_in1kz*seresnextaa101d_32x8d.sw_in12k_ft_in1k_288)é	   rf  )ra  r>  r<  r=  rP  rQ  rB  z&seresnextaa101d_32x8d.sw_in12k_ft_in1k)ra  rB  rQ  z*seresnextaa201d_32x8d.sw_in12k_ft_in1k_384rH  )é   rg  )rD   é€  rh  )ra  r?  rB  r=  r<  r>  zseresnextaa201d_32x8d.sw_in12ki-.  )	ra  rÏ   r?  rB  r>  r<  r=  rP  rQ  zresnetaa50d.sw_in12k)ra  rÏ   rB  r>  rQ  zresnetaa50d.d_in12kzresnetaa101d.sw_in12kzseresnextaa101d_32x8d.sw_in12kzresnetblur18.untrainedzresnetblur50.bt_in1kzjhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnetblur50-84f4748f.pthzresnetblur50d.untrainedzresnetblur101d.untrainedzresnetaa34d.untrainedzresnetaa50.a1h_in1kzphttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rsb-weights/resnetaa50_a1h-4cf422b3.pthzseresnetaa50d.untrainedzseresnextaa101d_32x8d.ah_in1kzzhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tpu-weights/seresnextaa101d_32x8d_ah-83c8ae12.pthzresnetrs50.tf_in1kzohttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs50_ema-6b53758b.pthrY  r[  g…ëQ¸í?)ra  r7  r<  r=  r>  rP  r?  rB  zresnetrs101.tf_in1kzuhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs101_i192_ema-1509bbf6.pth)rD   éÀ   ri  zresnetrs152.tf_in1kzuhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs152_i256_ema-a9aff7f9.pthzresnetrs200.tf_in1kzohttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-tpu-weights/resnetrs200_c-6b698b88.pthzresnetrs270.tf_in1kzphttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs270_ema-b40e674c.pthzresnetrs350.tf_in1kzuhttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs350_i256_ema-5a1aa8f1.pthzresnetrs420.tf_in1kzphttps://github.com/rwightman/pytorch-image-models/releases/download/v0.1-rs-weights/resnetrs420_ema-972dee69.pth)rD   é   rj  zresnet18.gluon_in1kzrhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet18_v1b-0757602b.pthzresnet34.gluon_in1kzrhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet34_v1b-c6d82d59.pthzresnet50.gluon_in1kzrhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet50_v1b-0ebe02e2.pthzresnet101.gluon_in1kzshttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet101_v1b-3b017079.pthzresnet152.gluon_in1kzshttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet152_v1b-c1edb0dd.pthzresnet50c.gluon_in1kzrhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet50_v1c-48092f55.pthzresnet101c.gluon_in1kzshttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet101_v1c-1f26822a.pthzresnet152c.gluon_in1kzshttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet152_v1c-a3bb0b98.pthzresnet50d.gluon_in1kzrhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet50_v1d-818a1b1b.pthzresnet101d.gluon_in1kzshttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet101_v1d-0f9c8644.pthzresnet152d.gluon_in1kzshttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet152_v1d-bd354e12.pthzresnet50s.gluon_in1kzrhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet50_v1s-1762acc0.pthzresnet101s.gluon_in1kzshttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet101_v1s-60fe0cc1.pthzresnet152s.gluon_in1kzshttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnet152_v1s-dcc41b81.pthzresnext50_32x4d.gluon_in1kzuhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnext50_32x4d-e6a097c1.pthzresnext101_32x4d.gluon_in1kzvhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnext101_32x4d-b253c8c4.pthzresnext101_64x4d.gluon_in1kzvhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_resnext101_64x4d-f9a8e184.pthzseresnext50_32x4d.gluon_in1kzwhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_seresnext50_32x4d-90cf2d6e.pthzseresnext101_32x4d.gluon_in1kzxhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_seresnext101_32x4d-cf52900d.pthzseresnext101_64x4d.gluon_in1kzxhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_seresnext101_64x4d-f9926f93.pthzsenet154.gluon_in1kznhttps://github.com/rwightman/pytorch-pretrained-gluonresnet/releases/download/v0.1/gluon_senet154-70a1a3c0.pthztest_resnet.r160_in1k)ra  r@  rA  r>  r<  r=  rB  c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-10-T model.
    ©r   r   r   r   r    Údeep_tieredT©rÍ   rÎ   rÒ   rÓ   r¥   Ú	resnet10t©r±   r#   r6  ©r4  rµ   Ú
model_argss      r-   ro  ro  Ç  óF   € õ �J¨|ÈÐVcÐnrÐsÑsÔs€JÝ˜+ zÐPÐPµT¸*Ð5OÐ5OÈÐ5OÐ5OÐPÐPÐPr/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-14-T model.
    rl  r    rm  Trn  Ú	resnet14t©r±   r$   r6  rq  s      r-   ru  ru  Ï  rs  r/   c           	      ó`   — t          t          d¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z"Constructs a ResNet-18 model.
    ©r*   r*   r*   r*   ©rÍ   rÎ   Úresnet18rp  rq  s      r-   rz  rz  ×  ó>   € õ �J¨|Ð<Ñ<Ô<€JÝ˜* jÐOÐOµD¸Ð4NÐ4NÀvÐ4NÐ4NÐOÐOÐOr/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-18-D model.
    rx  r    rÛ   Trn  Ú	resnet18drp  rq  s      r-   r}  r}  ß  óF   € õ �J¨|ÈÐV\ÐgkÐlÑlÔl€JÝ˜+ zÐPÐPµT¸*Ð5OÐ5OÈÐ5OÐ5OÐPÐPÐPr/   c           	      ó`   — t          t          d¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z"Constructs a ResNet-34 model.
    ©rD   r}   r`  rD   ry  Úresnet34rp  rq  s      r-   r�  r�  ç  r{  r/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-34-D model.
    r€  r    rÛ   Trn  Ú	resnet34drp  rq  s      r-   rƒ  rƒ  ï  r~  r/   c           	      ó`   — t          t          d¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z"Constructs a ResNet-26 model.
    rx  ry  Úresnet26rv  rq  s      r-   r…  r…  ÷  r{  r/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-26-T model.
    rx  r    rm  Trn  Ú	resnet26trv  rq  s      r-   r‡  r‡  ÿ  rs  r/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-26-D model.
    rx  r    rÛ   Trn  Ú	resnet26drv  rq  s      r-   r‰  r‰    r~  r/   c           	      ó`   — t          t          d¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z"Constructs a ResNet-50 model.
    r€  ry  Úresnet50rv  rq  s      r-   r‹  r‹    r{  r/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-50-C model.
    r€  r    rÛ   ©rÍ   rÎ   rÒ   rÓ   Ú	resnet50crv  rq  s      r-   rŽ  rŽ    óC   € õ �J¨|ÈÐV\Ð]Ñ]Ô]€JÝ˜+ zÐPÐPµT¸*Ð5OÐ5OÈÐ5OÐ5OÐPÐPÐPr/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-50-D model.
    r€  r    rÛ   Trn  Ú	resnet50drv  rq  s      r-   r‘  r‘    r~  r/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-50-S model.
    r€  r1   rÛ   r�  Ú	resnet50srv  rq  s      r-   r“  r“  '  r�  r/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z$Constructs a ResNet-50-T model.
    r€  r    rm  Trn  Ú	resnet50trv  rq  s      r-   r•  r•  /  rs  r/   c           	      ó`   — t          t          d¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z#Constructs a ResNet-101 model.
    ©rD   r}   é   rD   ry  Ú	resnet101rv  rq  s      r-   r™  r™  7  ó>   € õ �J¨}Ð=Ñ=Ô=€JÝ˜+ zÐPÐPµT¸*Ð5OÐ5OÈÐ5OÐ5OÐPÐPÐPr/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z%Constructs a ResNet-101-C model.
    r—  r    rÛ   r�  Ú
resnet101crv  rq  s      r-   rœ  rœ  ?  óC   € õ �J¨}ÈÐW]Ð^Ñ^Ô^€JÝ˜,¨
ÐQÐQµd¸:Ð6PÐ6PÈÐ6PÐ6PÐQÐQÐQr/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z%Constructs a ResNet-101-D model.
    r—  r    rÛ   Trn  Ú
resnet101drv  rq  s      r-   rŸ  rŸ  G  óF   € õ �J¨}ÈÐW]ÐhlÐmÑmÔm€JÝ˜,¨
ÐQÐQµd¸:Ð6PÐ6PÈÐ6PÐ6PÐQÐQÐQr/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z%Constructs a ResNet-101-S model.
    r—  r1   rÛ   r�  Ú
resnet101srv  rq  s      r-   r¢  r¢  O  r�  r/   c           	      ó`   — t          t          d¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z#Constructs a ResNet-152 model.
    ©rD   rÙ   é$   rD   ry  Ú	resnet152rv  rq  s      r-   r¦  r¦  W  rš  r/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z%Constructs a ResNet-152-C model.
    r¤  r    rÛ   r�  Ú
resnet152crv  rq  s      r-   r¨  r¨  _  r�  r/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z%Constructs a ResNet-152-D model.
    r¤  r    rÛ   Trn  Ú
resnet152drv  rq  s      r-   rª  rª  g  r   r/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z%Constructs a ResNet-152-S model.
    r¤  r1   rÛ   r�  Ú
resnet152srv  rq  s      r-   r¬  r¬  o  r�  r/   c           	      ó`   — t          t          d¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z#Constructs a ResNet-200 model.
    ©rD   é   r¥  rD   ry  Ú	resnet200rv  rq  s      r-   r°  r°  w  s>   € õ �J¨~Ð>Ñ>Ô>€JÝ˜+ zÐPÐPµT¸*Ð5OÐ5OÈÐ5OÐ5OÐPÐPÐPr/   c           	      óf   — t          t          dddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z%Constructs a ResNet-200-D model.
    r®  r    rÛ   Trn  Ú
resnet200drv  rq  s      r-   r²  r²    sF   € õ �J¨~È"ÐX^ÐimÐnÑnÔn€JÝ˜,¨
ÐQÐQµd¸:Ð6PÐ6PÈÐ6PÐ6PÐQÐQÐQr/   c           	      ób   — t          t          dd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )aO  Constructs a Wide ResNet-50-2 model.
    The model is the same as ResNet except for the bottleneck number of channels
    which is twice larger in every block. The number of channels in outer 1x1
    convolutions is the same, e.g. last block in ResNet-50 has 2048-512-2048
    channels, and in Wide ResNet-50-2 has 2048-1024-2048.
    r€  rË   ©rÍ   rÎ   r6   Úwide_resnet50_2rv  rq  s      r-   rµ  rµ  ‡  sA   € õ �J¨|ÈÐLÑLÔL€JÝÐ+¨ZÐVÐV½4À
Ð;UÐ;UÈfÐ;UÐ;UÐVÐVÐVr/   c           	      ób   — t          t          dd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )zçConstructs a Wide ResNet-101-2 model.
    The model is the same as ResNet except for the bottleneck number of channels
    which is twice larger in every block. The number of channels in outer 1x1
    convolutions is the same.
    r—  rË   r´  Úwide_resnet101_2rv  rq  s      r-   r·  r·  “  sA   € õ �J¨}ÈÐMÑMÔM€JÝÐ,¨jÐWÐW½DÀÐ<VÐ<VÈvÐ<VÐ<VÐWÐWÐWr/   c           	      ób   — t          t          dd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z.Constructs a ResNet-50 model w/ GroupNorm
    r€  Ú	groupnorm)rÍ   rÎ   r:   Úresnet50_gnrv  rq  s      r-   rº  rº  ž  s@   € õ �J¨|ÈÐTÑTÔT€JÝ˜-¨ÐRÐRµt¸JÐ7QÐ7QÈ&Ð7QÐ7QÐRÐRÐRr/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z(Constructs a ResNeXt50-32x4d model.
    r€  r    r}   ©rÍ   rÎ   r5   r6   Úresnext50_32x4drv  rq  s      r-   r½  r½  ¦  sD   € õ �J¨|ÈÐXYÐZÑZÔZ€JÝÐ+¨ZÐVÐV½4À
Ð;UÐ;UÈfÐ;UÐ;UÐVÐVÐVr/   c           	      ój   — t          t          dddddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )zVConstructs a ResNeXt50d-32x4d model. ResNext50 w/ deep stem & avg pool downsample
    r€  r    r}   rÛ   T)rÍ   rÎ   r5   r6   rÒ   rÓ   r¥   Úresnext50d_32x4drv  rq  s      r-   r¿  r¿  ®  sS   € õ Ý ¸BÈ1Ø °$ð8ñ 8ô 8€Jõ Ð,¨jÐWÐW½DÀÐ<VÐ<VÈvÐ<VÐ<VÐWÐWÐWr/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z*Constructs a ResNeXt-101 32x4d model.
    r—  r    r}   r¼  Úresnext101_32x4drv  rq  s      r-   rÁ  rÁ  ¸  óD   € õ �J¨}È"ÐYZÐ[Ñ[Ô[€JÝÐ,¨jÐWÐW½DÀÐ<VÐ<VÈvÐ<VÐ<VÐWÐWÐWr/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z*Constructs a ResNeXt-101 32x8d model.
    r—  r    rÙ   r¼  Úresnext101_32x8drv  rq  s      r-   rÄ  rÄ  À  rÂ  r/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z*Constructs a ResNeXt-101 32x16d model
    r—  r    rÚ   r¼  Úresnext101_32x16drv  rq  s      r-   rÆ  rÆ  È  óE   € õ �J¨}È"ÐY[Ð\Ñ\Ô\€JÝÐ-¨zÐXÐX½TÀ*Ð=WÐ=WÐPVÐ=WÐ=WÐXÐXÐXr/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z*Constructs a ResNeXt-101 32x32d model
    r—  r    r¼  Úresnext101_32x32drv  rq  s      r-   rÉ  rÉ  Ð  rÇ  r/   c           	      ód   — t          t          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z)Constructs a ResNeXt101-64x4d model.
    r—  r1   r}   r¼  Úresnext101_64x4drv  rq  s      r-   rË  rË  Ø  rÂ  r/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	z»Constructs an ECA-ResNeXt-26-T model.
    This is technically a 28 layer ResNet, like a 'D' bag-of-tricks model but with tiered 24, 32, 64 channels
    in the deep stem and ECA attn.
    rx  r    rm  TÚeca©r;   ©rÍ   rÎ   rÒ   rÓ   r¥   r×   Úecaresnet26trv  rq  s      r-   rÐ  rÐ  à  s^   € õ Ý ¸"Ø¨$½4È5Ð;QÑ;QÔ;QðSñ Sô S€Jõ ˜.¨*ÐSÐS½¸ZÐ8RÐ8RÈ6Ð8RÐ8RÐSÐSÐSr/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	z-Constructs a ResNet-50-D model with eca.
    r€  r    rÛ   TrÍ  rÎ  rÏ  Úecaresnet50drv  rq  s      r-   rÒ  rÒ  ì  s\   € õ Ý ¸"ÈÐY]Ý 5Ð)Ñ)Ô)ð+ñ +ô +€Jõ ˜.¨*ÐSÐS½¸ZÐ8RÐ8RÈ6Ð8RÐ8RÐSÐSÐSr/   c           
      óˆ   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fd	dit          |fi |¤Ž¤ŽS )
z…Constructs a ResNet-50-D model pruned with eca.
        The pruning has been obtained using https://arxiv.org/pdf/2002.08258.pdf
    r€  r    rÛ   TrÍ  rÎ  rÏ  Úecaresnet50d_prunedÚprunedrv  rq  s      r-   rÔ  rÔ  ö  sd   € õ
 Ý ¸"ÈÐY]Ý 5Ð)Ñ)Ô)ð+ñ +ô +€Jõ Ð/°ÐgÐgÀDÐgÍDÐQ[ÐLfÐLfÐ_eÐLfÐLfÐgÐgÐgr/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	z�Constructs an ECA-ResNet-50-T model.
    Like a 'D' bag-of-tricks model but with tiered 24, 32, 64 channels in the deep stem and ECA attn.
    r€  r    rm  TrÍ  rÎ  rÏ  Úecaresnet50trv  rq  s      r-   r×  r×    s^   € õ
 Ý ¸"Ø¨$½4È5Ð;QÑ;QÔ;QðSñ Sô S€Jõ ˜.¨*ÐSÐS½¸ZÐ8RÐ8RÈ6Ð8RÐ8RÐSÐSÐSr/   c           	      ó‚   — t          t          dddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z3Constructs a ResNet-50-D light model with eca.
    )r   r   é   rD   r    TrÍ  rÎ  )rÍ   rÎ   rÒ   r¥   r×   Úecaresnetlightrv  rq  s      r-   rÚ  rÚ    sZ   € õ Ý ¸2ÈÝ 5Ð)Ñ)Ô)ð+ñ +ô +€Jõ Ð*¨JÐUÐU½$¸zÐ:TÐ:TÈVÐ:TÐ:TÐUÐUÐUr/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	z.Constructs a ResNet-101-D model with eca.
    r—  r    rÛ   TrÍ  rÎ  rÏ  Úecaresnet101drv  rq  s      r-   rÜ  rÜ    s\   € õ Ý ¸2ÈÐZ^Ý 5Ð)Ñ)Ô)ð+ñ +ô +€Jõ ˜/¨:ÐTÐT½¸jÐ9SÐ9SÈFÐ9SÐ9SÐTÐTÐTr/   c           
      óˆ   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fd	dit          |fi |¤Ž¤ŽS )
z…Constructs a ResNet-101-D model pruned with eca.
       The pruning has been obtained using https://arxiv.org/pdf/2002.08258.pdf
    r—  r    rÛ   TrÍ  rÎ  rÏ  Úecaresnet101d_prunedrÕ  rv  rq  s      r-   rÞ  rÞ     sd   € õ
 Ý ¸2ÈÐZ^Ý 5Ð)Ñ)Ô)ð+ñ +ô +€Jõ Ð0°*ÐhÐhÀTÐhÍTÐR\ÐMgÐMgÐ`fÐMgÐMgÐhÐhÐhr/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	z.Constructs a ResNet-200-D model with ECA.
    r®  r    rÛ   TrÍ  rÎ  rÏ  Úecaresnet200drv  rq  s      r-   rà  rà  +  ó\   € õ Ý ¸BÈ&Ð[_Ý 5Ð)Ñ)Ô)ð+ñ +ô +€Jõ ˜/¨:ÐTÐT½¸jÐ9SÐ9SÈFÐ9SÐ9SÐTÐTÐTr/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	z.Constructs a ResNet-269-D model with ECA.
    ©rD   é   é0   rÙ   r    rÛ   TrÍ  rÎ  rÏ  Úecaresnet269drv  rq  s      r-   ræ  ræ  5  rá  r/   c                 óˆ   — t          t          ddddddt          d¬¦  «        ¬¦  «        }t          d	| fi t          |fi |¤Ž¤ŽS )
zàConstructs an ECA-ResNeXt-26-T model.
    This is technically a 28 layer ResNet, like a 'D' bag-of-tricks model but with tiered 24, 32, 64 channels
    in the deep stem. This model replaces SE module with the ECA module
    rx  r    r}   rm  TrÍ  rÎ  ©rÍ   rÎ   r5   r6   rÒ   rÓ   r¥   r×   Úecaresnext26t_32x4drv  rq  s      r-   ré  ré  ?  óe   € õ Ý ¸2È!ÐXZØ¨$½4È5Ð;QÑ;QÔ;QðSñ Sô S€Jõ Ð/°ÐZÐZ½tÀJÐ?YÐ?YÐRXÐ?YÐ?YÐZÐZÐZr/   c                 óˆ   — t          t          ddddddt          d¬¦  «        ¬¦  «        }t          d	| fi t          |fi |¤Ž¤ŽS )
zàConstructs an ECA-ResNeXt-50-T model.
    This is technically a 28 layer ResNet, like a 'D' bag-of-tricks model but with tiered 24, 32, 64 channels
    in the deep stem. This model replaces SE module with the ECA module
    rx  r    r}   rm  TrÍ  rÎ  rè  Úecaresnext50t_32x4drv  rq  s      r-   rì  rì  K  rê  r/   c           	      ó~   — t          t          dt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )Nrx  rW   rÎ  ©rÍ   rÎ   r×   Ú
seresnet18rp  rq  s      r-   rï  rï  W  óJ   € å�J¨|ÍÐX\ÐH]ÑH]ÔH]Ð^Ñ^Ô^€JÝ˜,¨
ÐQÐQµd¸:Ð6PÐ6PÈÐ6PÐ6PÐQÐQÐQr/   c           	      ó~   — t          t          dt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )Nr€  rW   rÎ  rî  Ú
seresnet34rp  rq  s      r-   rò  rò  ]  rð  r/   c           	      ó~   — t          t          dt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )Nr€  rW   rÎ  rî  Ú
seresnet50rv  rq  s      r-   rô  rô  c  rð  r/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	Nr€  r    rm  TrW   rÎ  rÏ  Úseresnet50trv  rq  s      r-   rö  rö  i  sY   € åÝ ¸2ÈØ¥$°$Ð"7Ñ"7Ô"7ð9ñ 9ô 9€Jõ ˜-¨ÐRÐRµt¸JÐ7QÐ7QÈ&Ð7QÐ7QÐRÐRÐRr/   c           	      ó~   — t          t          dt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )Nr—  rW   rÎ  rî  Úseresnet101rv  rq  s      r-   rø  rø  q  óJ   € å�J¨}ÍÐY]ÐI^ÑI^ÔI^Ð_Ñ_Ô_€JÝ˜-¨ÐRÐRµt¸JÐ7QÐ7QÈ&Ð7QÐ7QÐRÐRÐRr/   c           	      ó~   — t          t          dt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )Nr¤  rW   rÎ  rî  Úseresnet152rv  rq  s      r-   rû  rû  w  rù  r/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	Nr¤  r    rÛ   TrW   rÎ  rÏ  Úseresnet152drv  rq  s      r-   rý  rý  }  sY   € åÝ ¸2ÈØ¥$°$Ð"7Ñ"7Ô"7ð9ñ 9ô 9€Jõ ˜.¨*ÐSÐS½¸ZÐ8RÐ8RÈ6Ð8RÐ8RÐSÐSÐSr/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	z2Constructs a ResNet-200-D model with SE attn.
    r®  r    rÛ   TrW   rÎ  rÏ  Úseresnet200drv  rq  s      r-   rÿ  rÿ  …  ó[   € õ Ý ¸BÈ&Ø¥$°$Ð"7Ñ"7Ô"7ð9ñ 9ô 9€Jõ ˜.¨*ÐSÐS½¸ZÐ8RÐ8RÈ6Ð8RÐ8RÐSÐSÐSr/   c           
      ó„   — t          t          ddddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )	z2Constructs a ResNet-269-D model with SE attn.
    rã  r    rÛ   TrW   rÎ  rÏ  Úseresnet269drv  rq  s      r-   r  r  �  r   r/   c                 óˆ   — t          t          ddddddt          d¬¦  «        ¬¦  «        }t          d	| fi t          |fi |¤Ž¤ŽS )
zÃConstructs a SE-ResNeXt-26-D model.`
    This is technically a 28 layer ResNet, using the 'D' modifier from Gluon / bag-of-tricks for
    combination of deep stem and avg_pool in downsample.
    rx  r    r}   rÛ   TrW   rÎ  rè  Úseresnext26d_32x4drv  rq  s      r-   r  r  ™  se   € õ Ý ¸2È!ÐXZØ 4µDÀDÐ4IÑ4IÔ4IðKñ Kô K€Jõ Ð.°
ÐYÐY½dÀ:Ð>XÐ>XÐQWÐ>XÐ>XÐYÐYÐYr/   c                 óˆ   — t          t          ddddddt          d¬¦  «        ¬¦  «        }t          d	| fi t          |fi |¤Ž¤ŽS )
z«Constructs a SE-ResNet-26-T model.
    This is technically a 28 layer ResNet, like a 'D' bag-of-tricks model but with tiered 24, 32, 64 channels
    in the deep stem.
    rx  r    r}   rm  TrW   rÎ  rè  Úseresnext26t_32x4drv  rq  s      r-   r  r  ¥  se   € õ Ý ¸2È!ÐXZØ¨$½4È4Ð;PÑ;PÔ;PðRñ Rô R€Jõ Ð.°
ÐYÐY½dÀ:Ð>XÐ>XÐQWÐ>XÐ>XÐYÐYÐYr/   c           	      ó‚   — t          t          dddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )Nr€  r    r}   rW   rÎ  ©rÍ   rÎ   r5   r6   r×   Úseresnext50_32x4drv  rq  s      r-   r	  r	  ±  sY   € åÝ ¸2È!Ý 4Ð(Ñ(Ô(ð*ñ *ô *€Jõ Ð-¨zÐXÐX½TÀ*Ð=WÐ=WÐPVÐ=WÐ=WÐXÐXÐXr/   c           	      ó‚   — t          t          dddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )Nr—  r    r}   rW   rÎ  r  Úseresnext101_32x4drv  rq  s      r-   r  r  ¹  óY   € åÝ ¸BÈ1Ý 4Ð(Ñ(Ô(ð*ñ *ô *€Jõ Ð.°
ÐYÐY½dÀ:Ð>XÐ>XÐQWÐ>XÐ>XÐYÐYÐYr/   c           	      ó‚   — t          t          dddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )Nr—  r    rÙ   rW   rÎ  r  Úseresnext101_32x8drv  rq  s      r-   r  r  Á  r  r/   c                 óˆ   — t          t          ddddddt          d¬¦  «        ¬¦  «        }t          d	| fi t          |fi |¤Ž¤ŽS )
Nr—  r    rÙ   rÛ   TrW   rÎ  rè  Úseresnext101d_32x8drv  rq  s      r-   r  r  É  s`   € åÝ ¸BÈ1Ø °$Ý 4Ð(Ñ(Ô(ð*ñ *ô *€Jõ Ð/°ÐZÐZ½tÀJÐ?YÐ?YÐRXÐ?YÐ?YÐZÐZÐZr/   c           	      ó‚   — t          t          dddt          d¬¦  «        ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )Nr—  r1   r}   rW   rÎ  r  Úseresnext101_64x4drv  rq  s      r-   r  r  Ò  r  r/   c                 óˆ   — t          t          ddddddt          d¬¦  «        ¬	¦  «        }t          d
| fi t          |fi |¤Ž¤ŽS )Nr¤  r1   r}   rÛ   rD   r*   rW   rÎ  )rÍ   rÎ   r5   r6   rÓ   r¤   rÕ   r×   Úsenet154rv  rq  s      r-   r  r  Ú  sa   € åÝ ¸BÈ1ÐX^Ø¨q½TÈTÐ=RÑ=RÔ=RðTñ Tô T€Jõ ˜* jÐOÐOµD¸Ð4NÐ4NÀvÐ4NÐ4NÐOÐOÐOr/   c           	      ól   — t          t          dt          ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z9Constructs a ResNet-18 model with blur anti-aliasing
    rx  ©rÍ   rÎ   r<   Úresnetblur18)r±   r#   r   r6  rq  s      r-   r  r  â  ó@   € õ �J¨|ÅjÐQÑQÔQ€JÝ˜.¨*ÐSÐS½¸ZÐ8RÐ8RÈ6Ð8RÐ8RÐSÐSÐSr/   c           	      ól   — t          t          dt          ¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z9Constructs a ResNet-50 model with blur anti-aliasing
    r€  r  Úresnetblur50©r±   r$   r   r6  rq  s      r-   r  r  ê  r  r/   c           	      ór   — t          t          dt          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z;Constructs a ResNet-50-D model with blur anti-aliasing
    r€  r    rÛ   T©rÍ   rÎ   r<   rÒ   rÓ   r¥   Úresnetblur50dr  rq  s      r-   r  r  ò  sP   € õ Ý ½
Ø °$ð8ñ 8ô 8€Jõ ˜/¨:ÐTÐT½¸jÐ9SÐ9SÈFÐ9SÐ9SÐTÐTÐTr/   c           	      ór   — t          t          dt          ddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z<Constructs a ResNet-101-D model with blur anti-aliasing
    r—  r    rÛ   Tr  Úresnetblur101dr  rq  s      r-   r   r   ü  sQ   € õ Ý ½Ø °$ð8ñ 8ô 8€Jõ Ð*¨JÐUÐU½$¸zÐ:TÐ:TÈVÐ:TÐ:TÐUÐUÐUr/   c           	      ó|   — t          t          dt          j        ddd¬¦  «        }t	          d| fi t          |fi |¤Ž¤ŽS )z<Constructs a ResNet-34-D model w/ avgpool anti-aliasing
    r€  r    rÛ   Tr  Úresnetaa34d)r±   r#   rN   r’   r6  rq  s      r-   r"  r"    sU   € õ Ý ½¼ÐRTÐ`fÐquðwñ wô w€Jå˜-¨ÐRÐRµt¸JÐ7QÐ7QÈ&Ð7QÐ7QÐRÐRÐRr/   c           	      óv   — t          t          dt          j        ¬¦  «        }t	          d| fi t          |fi |¤Ž¤ŽS )z<Constructs a ResNet-50 model with avgpool anti-aliasing
    r€  r  Ú
resnetaa50©r±   r$   rN   r’   r6  rq  s      r-   r$  r$    sB   € õ �J¨|ÅbÄlÐSÑSÔS€JÝ˜,¨
ÐQÐQµd¸:Ð6PÐ6PÈÐ6PÐ6PÐQÐQÐQr/   c           	      ó|   — t          t          dt          j        ddd¬¦  «        }t	          d| fi t          |fi |¤Ž¤ŽS )z>Constructs a ResNet-50-D model with avgpool anti-aliasing
    r€  r    rÛ   Tr  Úresnetaa50dr%  rq  s      r-   r'  r'    sR   € õ Ý ½¼Ø °$ð8ñ 8ô 8€Jõ ˜-¨ÐRÐRµt¸JÐ7QÐ7QÈ&Ð7QÐ7QÐRÐRÐRr/   c           	      ó|   — t          t          dt          j        ddd¬¦  «        }t	          d| fi t          |fi |¤Ž¤ŽS )z?Constructs a ResNet-101-D model with avgpool anti-aliasing
    r—  r    rÛ   Tr  Úresnetaa101dr%  rq  s      r-   r)  r)  !  sR   € õ Ý ½¼Ø °$ð8ñ 8ô 8€Jõ ˜.¨*ÐSÐS½¸ZÐ8RÐ8RÈ6Ð8RÐ8RÐSÐSÐSr/   c                 óš   — t          t          dt          j        dddt          d¬¦  «        ¬¦  «        }t	          d| fi t          |fi |¤Ž¤ŽS )	zAConstructs a SE=ResNet-50-D model with avgpool anti-aliasing
    r€  r    rÛ   TrW   rÎ  )rÍ   rÎ   r<   rÒ   rÓ   r¥   r×   Úseresnetaa50dr%  rq  s      r-   r+  r+  +  sc   € õ Ý ½¼Ø °$Å4ÐSWÐCXÑCXÔCXðZñ Zô Z€Jõ ˜/¨:ÐTÐT½¸jÐ9SÐ9SÈFÐ9SÐ9SÐTÐTÐTr/   c                 óž   — t          t          ddddddt          j        t          d¬¦  «        ¬¦	  «	        }t	          d	| fi t          |fi |¤Ž¤ŽS )
úIConstructs a SE=ResNeXt-101-D 32x8d model with avgpool anti-aliasing
    r—  r    rÙ   rÛ   TrW   rÎ  ©	rÍ   rÎ   r5   r6   rÒ   rÓ   r¥   r<   r×   Úseresnextaa101d_32x8dr%  rq  s      r-   r/  r/  5  sf   € õ Ý ¸BÈ1Ø °$ÅÄÝ 4Ð(Ñ(Ô(ð*ñ *ô *€Jõ Ð1°:Ð\Ð\ÅÀjÐA[ÐA[ÐTZÐA[ÐA[Ð\Ð\Ð\r/   c                 óž   — t          t          ddddddt          j        t          d¬¦  «        ¬	¦	  «	        }t	          d
| fi t          |fi |¤Ž¤ŽS )r-  )rD   r¯  r¥  r}   r    rÙ   r1   rÛ   TrW   rÎ  r.  Úseresnextaa201d_32x8dr%  rq  s      r-   r1  r1  @  sf   € õ Ý ¸RÈAØ °$ÅÄÝ 4Ð(Ñ(Ô(ð*ñ *ô *€Jõ Ð1°:Ð\Ð\ÅÀjÐA[ÐA[ÐTZÐA[ÐA[Ð\Ð\Ð\r/   c                 óÂ   — t          t          d¦  «        d¬¦  «        }t          t          dddddt          |¬¦  «        ¬	¦  «        }t	          d
| fi t          |fi |¤Ž¤ŽS )zÓConstructs a ResNet-RS-50 model.
    Paper: Revisiting ResNets - https://arxiv.org/abs/2103.07579
    Pretrained weights from https://github.com/tensorflow/tpu/tree/bee9c4f6/models/official/resnet/resnet_rs
    rW   r›   ©Úrd_ratior€  r    rÛ   TrÎ  ©rÍ   rÎ   rÒ   rÓ   rÔ   r¥   r×   Ú
resnetrs50©r   r   r±   r$   r6  ©r4  rµ   r;   rr  s       r-   r6  r6  K  sy   € õ � $™œ°$Ð7Ñ7Ô7€JÝÝ ¸"ÈÐbfØ¥4°:Ð#>Ñ#>Ô#>ð@ñ @ô @€Jõ ˜,¨
ÐQÐQµd¸:Ð6PÐ6PÈÐ6PÐ6PÐQÐQÐQr/   c                 óÂ   — t          t          d¦  «        d¬¦  «        }t          t          dddddt          |¬¦  «        ¬	¦  «        }t	          d
| fi t          |fi |¤Ž¤ŽS )zÔConstructs a ResNet-RS-101 model.
    Paper: Revisiting ResNets - https://arxiv.org/abs/2103.07579
    Pretrained weights from https://github.com/tensorflow/tpu/tree/bee9c4f6/models/official/resnet/resnet_rs
    rW   r›   r3  r—  r    rÛ   TrÎ  r5  Úresnetrs101r7  r8  s       r-   r:  r:  X  óy   € õ � $™œ°$Ð7Ñ7Ô7€JÝÝ ¸2ÈÐcgØ¥4°:Ð#>Ñ#>Ô#>ð@ñ @ô @€Jõ ˜-¨ÐRÐRµt¸JÐ7QÐ7QÈ&Ð7QÐ7QÐRÐRÐRr/   c                 óÂ   — t          t          d¦  «        d¬¦  «        }t          t          dddddt          |¬¦  «        ¬	¦  «        }t	          d
| fi t          |fi |¤Ž¤ŽS )zÔConstructs a ResNet-RS-152 model.
    Paper: Revisiting ResNets - https://arxiv.org/abs/2103.07579
    Pretrained weights from https://github.com/tensorflow/tpu/tree/bee9c4f6/models/official/resnet/resnet_rs
    rW   r›   r3  r¤  r    rÛ   TrÎ  r5  Úresnetrs152r7  r8  s       r-   r=  r=  e  r;  r/   c                 óÂ   — t          t          d¦  «        d¬¦  «        }t          t          dddddt          |¬¦  «        ¬	¦  «        }t	          d
| fi t          |fi |¤Ž¤ŽS )zÔConstructs a ResNet-RS-200 model.
    Paper: Revisiting ResNets - https://arxiv.org/abs/2103.07579
    Pretrained weights from https://github.com/tensorflow/tpu/tree/bee9c4f6/models/official/resnet/resnet_rs
    rW   r›   r3  r®  r    rÛ   TrÎ  r5  Úresnetrs200r7  r8  s       r-   r?  r?  r  óy   € õ � $™œ°$Ð7Ñ7Ô7€JÝÝ ¸BÈ&ÐdhØ¥4°:Ð#>Ñ#>Ô#>ð@ñ @ô @€Jõ ˜-¨ÐRÐRµt¸JÐ7QÐ7QÈ&Ð7QÐ7QÐRÐRÐRr/   c                 óÂ   — t          t          d¦  «        d¬¦  «        }t          t          dddddt          |¬¦  «        ¬	¦  «        }t	          d
| fi t          |fi |¤Ž¤ŽS )zÔConstructs a ResNet-RS-270 model.
    Paper: Revisiting ResNets - https://arxiv.org/abs/2103.07579
    Pretrained weights from https://github.com/tensorflow/tpu/tree/bee9c4f6/models/official/resnet/resnet_rs
    rW   r›   r3  )r}   é   é5   r}   r    rÛ   TrÎ  r5  Úresnetrs270r7  r8  s       r-   rD  rD    r@  r/   c                 óÂ   — t          t          d¦  «        d¬¦  «        }t          t          dddddt          |¬¦  «        ¬	¦  «        }t	          d
| fi t          |fi |¤Ž¤ŽS )zÔConstructs a ResNet-RS-350 model.
    Paper: Revisiting ResNets - https://arxiv.org/abs/2103.07579
    Pretrained weights from https://github.com/tensorflow/tpu/tree/bee9c4f6/models/official/resnet/resnet_rs
    rW   r›   r3  )r}   r¥  éH   r}   r    rÛ   TrÎ  r5  Úresnetrs350r7  r8  s       r-   rG  rG  �  r@  r/   c                 óÂ   — t          t          d¦  «        d¬¦  «        }t          t          dddddt          |¬¦  «        ¬	¦  «        }t	          d
| fi t          |fi |¤Ž¤ŽS )zÓConstructs a ResNet-RS-420 model
    Paper: Revisiting ResNets - https://arxiv.org/abs/2103.07579
    Pretrained weights from https://github.com/tensorflow/tpu/tree/bee9c4f6/models/official/resnet/resnet_rs
    rW   r›   r3  )r}   é,   éW   r}   r    rÛ   TrÎ  r5  Úresnetrs420r7  r8  s       r-   rK  rK  š  r@  r/   c           	      óŽ   — t          t          t          t          t          gddddd¬¦  «        }t          d| fi t          |fi |¤Ž¤ŽS )z)Constructs a tiny ResNet test model.
    rl  rÚ   rÛ   T)r    rå  rå  é`   )rÍ   rÎ   rÒ   rÓ   r¥   rI   Útest_resnet)r±   r#   r$   r6  rq  s      r-   rN  rN  §  s]   € õ Ý�:¥zµ:Ð>À|Ø °$ÐAQðSñ Sô S€Jõ ˜-¨ÐRÐRµt¸JÐ7QÐ7QÈ&Ð7QÐ7QÐRÐRÐRr/   Útv_resnet34Útv_resnet50Útv_resnet101Útv_resnet152Útv_resnext50_32x4dÚig_resnext101_32x8dÚig_resnext101_32x16dÚig_resnext101_32x32dÚig_resnext101_32x48dÚssl_resnet18Ússl_resnet50Ússl_resnext50_32x4dÚssl_resnext101_32x4dÚssl_resnext101_32x8dÚssl_resnext101_32x16dÚswsl_resnet18Úswsl_resnet50Úswsl_resnext50_32x4dÚswsl_resnext101_32x4dÚswsl_resnext101_32x8dÚswsl_resnext101_32x16dÚgluon_resnet18_v1bÚgluon_resnet34_v1bÚgluon_resnet50_v1bÚgluon_resnet101_v1bÚgluon_resnet152_v1bÚgluon_resnet50_v1cÚgluon_resnet101_v1cÚgluon_resnet152_v1cÚgluon_resnet50_v1dÚgluon_resnet101_v1dÚgluon_resnet152_v1dÚgluon_resnet50_v1sÚgluon_resnet101_v1sr  )	Úgluon_resnet152_v1sÚgluon_resnext50_32x4dÚgluon_resnext101_32x4dÚgluon_resnext101_64x4dÚgluon_seresnext50_32x4dÚgluon_seresnext101_32x4dÚgluon_seresnext101_64x4dÚgluon_senet154Úseresnext26tn_32x4d)r   )r   r   NNNN)r—   )r   r    r   Fr—   r—   NNr/  )rÊ   )•rs   r€   Ú	functoolsr   Útypingr   r   r   r   r   r	   r
   rx   Útorch.nnrN   Útorch.nn.functionalÚ
functionalr*  Ú	timm.datar   r   Útimm.layersr   r   r   r   r   r   r   r   r   r   r   r   r   Ú_builderr   Ú	_featuresr   Ú_manipulater   Ú	_registryr   r    r!   Ú__all__rv   r.   rw   r#   r$   r�   r–   r,  rŸ   r1  r0  rÆ   r"   r6  rF  rJ  rS  rW  r\  r^  Údefault_cfgsro  ru  rz  r}  r�  rƒ  r…  r‡  r‰  r‹  rŽ  r‘  r“  r•  r™  rœ  rŸ  r¢  r¦  r¨  rª  r¬  r°  r²  rµ  r·  rº  r½  r¿  rÁ  rÄ  rÆ  rÉ  rË  rÐ  rÒ  rÔ  r×  rÚ  rÜ  rÞ  rà  ræ  ré  rì  rï  rò  rô  rö  rø  rû  rý  rÿ  r  r  r  r	  r  r  r  r  r  r  r  r  r   r"  r$  r'  r)  r+  r/  r1  r6  r:  r=  r?  rD  rG  rK  rN  rp   r+   r/   r-   ú<module>r‡     s?8  ððð ð €€€Ø Ð Ð Ð Ð Ð Ø @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 Aðgð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gð gà *Ð *Ð *Ð *Ð *Ð *Ø +Ð +Ð +Ð +Ð +Ð +Ø 'Ð 'Ð 'Ð 'Ð 'Ð 'Ø YÐ YÐ YÐ YÐ YÐ YÐ YÐ YÐ YÐ Yà
0Ð
0Ð
0€ñð ˜Sð ¨#ð ¸ð ÀSð ð ð ð ð
uð uð uð uð u�”ñ uô uð uðpuð uð uð uð u�”ñ uô uð uðx ØØ(,Ø04ØØñð Øðàðð ðð ð	ð
 ðð ! œðð ˜T "¤)œ_Ô-ðð „Yðð ð ð ðH ØØ(,Ø04ØØñð Øðàðð ðð ð	ð
 ðð ! œðð ˜T "¤)œ_Ô-ðð „Yðð ð ð ñ6jð j˜5ð j¨$¨x¸Ô/@Ô*Að jð jð jð jð( ØØ !ØØ!#Ø "ØØñV ð V Ø˜˜t JÔ/°°jÔ1AÐAÔBÀCÐGÔHðV à˜˜S˜”/ðV ð ˜S #˜X”ðV ð ð	V ð
 ðV ð ðV ð ðV ð ðV ð ðV ð ðV ð ˆ4��c˜2œ9�nÔ%Ô&¨¨T°#°s°(¬^Ô(<Ð<Ô=ðV ð V ð V ð V ðrXð Xð Xð Xð XˆRŒYñ Xô Xð Xñv
Gð G˜Cð G¨Tð GÈð Gð Gð Gð Gñ
ð 
ˆcð 
 T¨#¨s¨(¤^ð 
ð 
ð 
ð 
ñIð Iˆsð I d¨3°¨8¤nð Ið Ið Ið Iñ
ð �ð  t¨C°¨H¤~ð ð ð ð ñð ˆsð  d¨3°¨8¤nð ð ð ð ñð �ð  t¨C°¨H¤~ð ð ð ð ñð ˆsð  d¨3°¨8¤nð ð ð ð ð %Ð$ð s	&à˜˜Øð AØ ¨FÀ$ÐXeØð	ñ ô ðs	&ð ˜˜Øð AØ ¨FÀ$ÐXeØð	ñ ô ðs	&ð ˜˜ØØðAñ Aô Aðs	&ð ˜˜ØØðAñ Aô Aðs	&ð$ ˜˜ØØðAñ Aô Að%s	&ð* ˜&˜&ØØyØðñ ô ð+s	&ð2 $ U UØØ /¸CÈIð&Wñ &Wô &Wð3s	&ð8 ˜˜ØØðAñ Aô Að9s	&ð> ˜˜ØØðAñ Aô Að?s	&ðD ˜˜ØØØðñ ô ðEs	&ðL ˜˜ØØtðvñ vô vðMs	&ðR # E EØØ /¸Cð%Añ %Aô %AðSs	&ðX ˜&˜&ØØyØðñ ô ðYs	&ð` ˜˜ØØtðvñ vô vðas	&ðf ˜˜ØØuØðñ ô ðgs	&ðn ˜&˜&Øð CØ¨À&Ø }ÀCð	Iñ Iô Iðos	&ðx ˜˜ØØ}ðñ ô ðys	&ð s	&ð~ ˜˜Øð DØ ¨F¸SÐR_Ðorðtñ tô tðs	&ðF ˜˜ØØðAñ Aô AðGs	&ðL ˜˜ØØðAñ Aô AðMs	&ðR ˜˜ØØ~ð@ñ @ô @ðSs	&ðX ˜˜ØØ~ð@ñ @ô @ðYs	&ð^ ˜˜ØØ}ðñ ô ð_s	&ðd ˜˜ØØ}ðñ ô ðes	&ðj �u�uØØ|ð~ñ ~ô ~ðks	&ðp ˜˜ØØzð|ñ |ô |ðqs	&ðv ˜˜ØØyð{ñ {ô {ðws	&ð| ˜˜ØØyð{ñ {ô {ð}s	&ðB ˜˜ØØyð{ñ {ô {ðCs	&ðH ˜&˜&ØØyØðñ ô ðIs	&ðP $ U UØØ /Ø }ÀCØð	&ñ &ô &ðQs	&ðZ ˜˜Øð AØðñ ô ð[s	&ðb ˜˜Øð AØðñ ô ðcs	&ðj ˜˜Øð AØðñ ô ðks	&ð s	&ð s	&ðr ˜6˜6¨YÐ7Ñ7Ô7ðss	&ðt ˜%˜%ØØ}ðñ ô ðus	&ðz ˜˜Øð AðBñ Bô Bð{s	&ð@ ˜˜Øð AðBñ Bô BðAs	&ðF ˜˜Øð AðBñ Bô BðGs	&ðL ˜6˜6ØØzØ¨À&ÐSWØ¨=ð	:ñ :ô :ðMs	&ðV ˜%˜%ØØ}ðñ ô ðWs	&ð\ ˜˜Øð AðBñ Bô Bð]s	&ðb ˜˜Øð AðBñ Bô Bðcs	&ðh ˜˜Øð AðBñ Bô Bðis	&ðn ˜6˜6ØØzØ¨À&ÐSWØ¨=ð	:ñ :ô :ðos	&ðx ˜6˜6™8œ8ðys	&ðz ˜6˜6ØØzØ¨À&ÐSWØ¨=ð	:ñ :ô :ð{s	&ðD    ØØ~ð"@ñ "@ô "@ðEs	&ðN ˜˜ØØGØÐ+NðPñ Pô PðOs	&ðV ˜˜ØØGØÐ+NðPñ Pô PðWs	&ð^ ˜˜ØØGØÐ+NðPñ Pô Pð_s	&ð s	&ð s	&ðf ˜˜ØØGØ ¨FÀMÐafØÐ+Nð	Pñ Pô Pðgs	&ðp ˜˜ØØHØÐ+NðPñ Pô Pðqs	&ðx ˜$˜$ØØHØ ¨FÀMÐafØÐ+Nð	Pñ Pô Pðys	&ðB ˜˜ØØHØÐ+NðPñ Pô PðCs	&ðJ ˜$˜$ØØHØ ¨FÀMÐafØÐ+Nð	Pñ Pô PðKs	&ðT ˜t˜tØØNØÐ+Nð Pñ  Pô  PðUs	&ð\   ØØNØ ¨FÀMÐafØÐ+Nð	!Pñ !Pô !Pð]s	&ðf   ØØOØÐ+Nð!Pñ !Pô !Pðgs	&ðn    ØØOØ ¨FÀMÐafØÐ+Nð	"Pñ "Pô "Pðos	&ð| ˜F˜FØð AØðñ ô ð}s	&ðH   Øð Dð!Eñ !Eô !EðIs	&ðN ˜u˜uØð Gð Hñ  Hô  HðOs	&ðT ˜u˜uØð Gð Hñ  Hô  HðUs	&ðZ ˜v˜vØð Gð Hñ  Hô  Hð[s	&ð` ˜v˜vØð Að Bñ  Bô  Bðas	&ðf   ØØ|Øð!ñ !ô !ðgs	&ðn ! & &¡(¤(ðos	&ð s	&ð s	&ðp   Øð Cð!Dñ !Dô !Dðqs	&ðz ˜t˜tØØNØÐ+Nð Pñ  Pô  Pð{s	&ðB   ØØOØÐ+Nð!Pñ !Pô !PðCs	&ðJ   ØØOØÐ+Nð!Pñ !Pô !PðKs	&ðR   ØØNØ ¨FÀMÐafØÐ+Nð	!Pñ !Pô !PðSs	&ð\    ØØOØ ¨FÀMÐafØÐ+Nð	"Pñ "Pô "Pð]s	&ðn +¨D¨DØØQØÐ+[ð-]ñ -]ô -]ðos	&ðv ,¨T¨TØØRØÐ+[ð.]ñ .]ô .]ðws	&ð~ ,¨T¨TØØRØÐ+[ð.]ñ .]ô .]ðs	&ðF	 ,¨T¨TØØRØÐ+[ð.]ñ .]ô .]ðG	s	&ðT	 '¨¨ØØrØÐ+rð*tñ *tô *tðU	s	&ð\	 '¨¨ØØrØÐ+rð*tñ *tô *tð]	s	&ðd	 .¨t¨tØØxØÐ+rð0tñ 0tô 0tðe	s	&ðl	 /°°ØØyØÐ+rð1tñ 1tô 1tðm	s	&ðt	 /°°ØØyØÐ+rð1tñ 1tô 1tðu	s	&ð|	 0°°ØØzØÐ+rð2tñ 2tô 2tð}	s	&ðJ
 $ T TØØyØÐ+rð&tñ &tô &tðK
s	&ð s	&ð s	&ðR
 $ T TØÙyØÐ+rð&tñ &tô &tðS
s	&ñZ
 +¨D¨DØÙØÐ+rð-tñ -tô -tð[
s	&ñb
 ,¨T¨TØñ AØÐ+rð.tñ .tô .tðc
s	&ñj
 ,¨T¨TØñ AØÐ+rð.tñ .tô .tðk
s	&ñr
 -¨d¨dØñ BØÐ+rð/tñ /tô /tðs
s	&ñ~
 ˜V˜VØÙ|Ø¨À&Ø¨Mñ	;ñ ;ô ;ð
s	&ñH   ØÙzØ¨Mñ!;ñ !;ô !;ðIs	&ñP ˜e˜eØÙxØ¨DÀ-ñQñ Qô QðQs	&ñX $ U UØÙzØ¨DÀ-ñ&Qñ &Qô &QðYs	&ñ` ˜U˜UØÙ|Ø¨À&Ø¨Mñ	;ñ ;ô ;ðas	&ñj ˜E˜EØñ DØðñ ô ðks	&ñr ˜E˜EØñ DØðñ ô ðss	&ñz ˜F˜FØñ DØðñ ô ð{s	&ñB ˜u˜uØÙyØ¨DÀ-ñ Qñ  Qô  QðCs	&ñJ % e eØÙ{Ø¨DÀ-ñ'Qñ 'Qô 'QðKs	&ñR ˜v˜vØ¨ÀÐQWñ Yñ  Yô  YðSs	&ñV ˜f˜fØñ BØ¨Á(ÐUYØ©=ð	:ñ :ô :ðWs	&ð s	&ð s	&ñd $ U U°iÐ%@Ñ%@Ô%@ðes	&ñf $ U U°iÐ%@Ñ%@Ô%@ðgs	&ñl ˜F˜F™HœHðms	&ñn ˜F˜F™HœHðos	&ñp ˜%˜%Øñ BØðñ ô ðqs	&ñx ˜%˜%Øñ BØðñ ô ðys	&ñ@ ˜&˜&Øñ BØðñ ô ðAs	&ñH ˜6˜6ØÙ}ðñ ô ðIs	&ñN ˜V˜VØðñ ô ðOs	&ñR ˜V˜V™XœXðSs	&ñT ˜V˜V™XœXðUs	&ñV ˜V˜VØÙ|Ø¨À&ÐSWØ¨=ð	ñ ô ðWs	&ñb ˜f˜fØ¨À&ñJñ Jô Jðcs	&ñf ˜f˜fØ¨À&ñJñ Jô Jðgs	&ñn ! & &ØÙ~Øð#ñ #ô #ðos	&ñv ! & &ØÙØð#ñ #ô #ðws	&ñ~ " 6 6Øñ Cð$Dñ $Dô $Dðs	&ð s	&ð s	&ñD # F F¡H¤HðEs	&ñF ! % %Øñ Fð#Gñ #Gô #GðGs	&ñL " 5 5Øñ GØð$ñ $ô $ðMs	&ñX # F FØØ t¸3ñ%@ñ %@ô %@ðYs	&ñ^ $ V VØØ t¸3ñ&@ñ &@ô &@ð_s	&ñd 1°&°&ØØ -¹6ÐS`ÐpsØñ3ñ 3ô 3ðes	&ñl -¨f¨fØØ¨Cñ/1ñ /1ô /1ðms	&ñr 1°$°$ØÙ¨IÁÑVcÐnqñ3sñ 3sô 3sðss	&ñx % d dØÙ©¸yØ -¹8ÑUbÐruñ'wñ 'wô 'wðys	&ñB ˜F˜FØÙ i¸$ÈcñSñ Sô SðCs	&ñH ˜6˜6ØÙ i¸$ÈcñSñ Sô SðIs	&ñN ˜V˜VØÙ i¸$ÈcñSñ Sô SðOs	&ñT % f fØÙ i¸$Ècñ'Sñ 'Sô 'SðUs	&ñ\ ˜f˜f™hœhð]s	&ñ^ ˜F˜FØÙxðzñ zô zð_s	&ñd ˜v˜v°Ð;Ñ;Ô;ðes	&ñf   °)Ð <Ñ <Ô <ðgs	&ð s	&ð s	&ñh ˜V˜V¨yÐ9Ñ9Ô9ðis	&ñj ˜5˜5ØÙ~ð@ñ @ô @ðks	&ñr ˜v˜v°Ð;Ñ;Ô;ðss	&ñt $ U UØñ IØð&ñ &ô &ðus	&ñ@ ˜$˜$ØÙ}Ù ©F¹TÐS`Ù¨Iñ	7ñ 7ô 7ðAs	&ñJ ˜4˜4Øñ DÙ ¨F¸TÐS`Ù¨Iñ	7ñ 7ô 7ðKs	&ñT ˜4˜4Øñ DØ ¨F¸SÐR_Ù¨Iñ	7ñ 7ô 7ðUs	&ñ^ ˜4˜4ØÙ}Ø ¨F¸SÐR_Ù¨Iñ	7ñ 7ô 7ð_s	&ñh ˜4˜4ØÙ~Ø ¨F¸SÑR_Ù¨Iñ	7ñ 7ô 7ðis	&ñr ˜4˜4Øñ DØ ©F¸SÑR_Ù¨Iñ	7ñ 7ô 7ðss	&ñ| ˜4˜4ØÙ~Ø ©H¸sÑTaÙ¨Iñ	7ñ 7ô 7ð}s	&ñJ ˜5˜5Øñ AðBñ Bô BðKs	&ñP ˜5˜5Øñ AðBñ Bô BðQs	&ñV ˜5˜5Øñ AðBñ Bô BðWs	&ñ\ ˜E˜EØñ BðCñ Cô Cð]s	&ñb ˜E˜EØñ BðCñ Cô Cðcs	&ñh ˜E˜EØñ AØðñ ô ðis	&ð s	&ð s	&ñp ˜U˜UØñ BØðñ ô ðqs	&ñx ˜U˜UØñ BØðñ ô ðys	&ñ@ ˜E˜EØñ AØðñ ô ðAs	&ñH ˜U˜UØñ BØðñ ô ðIs	&ñP ˜U˜UØñ BØðñ ô ðQs	&ñX ˜E˜EØñ AØðñ ô ðYs	&ñ` ˜U˜UØñ BØðñ ô ðas	&ñh ˜U˜UØñ BØðñ ô ðis	&ñp ! % %Øñ Dð#Eñ #Eô #Eðqs	&ñv " 5 5Øñ Eð$Fñ $Fô $Fðws	&ñ| " 5 5Øñ Eð$Fñ $Fô $Fð}s	&ñB # E EØñ Fð%Gñ %Gô %GðCs	&ñH $ U UØñ Gð&Hñ &Hô &HðIs	&ñN $ U UØñ Gð&Hñ &Hô &HðOs	&ñT ˜5˜5ØÙ|Øðñ ô ðUs	&ñ^ ˜T˜TØØ /¸DÙ ©F¸yñJñ Jô Jð_s	&ð s	&ñ s	ô s	€ðl ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñPð P˜ð P°Fð Pñ Pð Pñ „ðPð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñPð P˜ð P°Fð Pñ Pð Pñ „ðPð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñPð P˜ð P°Fð Pñ Pð Pñ „ðPð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñPð P˜ð P°Fð Pñ Pð Pñ „ðPð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð ñQð Q˜$ð Q°Vð Qñ Qð Qñ „ðQð ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð ñWð W ð W¸6ð Wñ Wð Wñ „ðWð ñXð X ð X¸Fð Xñ Xð Xñ „ðXð ñSð S˜Dð S°vð Sñ Sð Sñ „ðSð ñWð W ð W¸6ð Wñ Wð Wñ „ðWð ñXð X ð X¸Fð Xñ Xð Xñ „ðXð ñXð X ð X¸Fð Xñ Xð Xñ „ðXð ñXð X ð X¸Fð Xñ Xð Xñ „ðXð ñYð Y $ð Y¸Vð Yñ Yð Yñ „ðYð ñYð Y $ð Y¸Vð Yñ Yð Yñ „ðYð ñXð X ð X¸Fð Xñ Xð Xñ „ðXð ñTð T˜Tð T¸ð Tñ Tð Tñ „ðTð ñTð T˜Tð T¸ð Tñ Tð Tñ „ðTð ñhð h Dð h¸vð hñ hð hñ „ðhð ñTð T˜Tð T¸ð Tñ Tð Tñ „ðTð ñVð V˜tð V¸&ð Vñ Vð Vñ „ðVð ñUð U˜dð U¸ð Uñ Uð Uñ „ðUð ñið i Tð iÀð iñ ið iñ „ðið ñUð U˜dð U¸ð Uñ Uð Uñ „ðUð ñUð U˜dð U¸ð Uñ Uð Uñ „ðUð ñ[ð [ Dð [¸vð [ñ [ð [ñ „ð[ð ñ[ð [ Dð [¸vð [ñ [ð [ñ „ð[ð ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð
 ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð
 ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð
 ñSð S˜Dð S°vð Sñ Sð Sñ „ðSð ñSð S˜Dð S°vð Sñ Sð Sñ „ðSð
 ñSð S˜Dð S°vð Sñ Sð Sñ „ðSð
 ñTð T˜Tð T¸ð Tñ Tð Tñ „ðTð ñTð T˜Tð T¸ð Tñ Tð Tñ „ðTð ñTð T˜Tð T¸ð Tñ Tð Tñ „ðTð ñZð Z 4ð Z¸fð Zñ Zð Zñ „ðZð ñZð Z 4ð Z¸fð Zñ Zð Zñ „ðZð ñYð Y $ð Y¸Vð Yñ Yð Yñ „ðYð ñZð Z 4ð Z¸fð Zñ Zð Zñ „ðZð ñZð Z 4ð Z¸fð Zñ Zð Zñ „ðZð ñ[ð [ Dð [¸vð [ñ [ð [ñ „ð[ð ñZð Z 4ð Z¸fð Zñ Zð Zñ „ðZð ñPð P˜ð P°Fð Pñ Pð Pñ „ðPð ñTð T˜Tð T¸ð Tñ Tð Tñ „ðTð ñTð T˜Tð T¸ð Tñ Tð Tñ „ðTð ñUð U˜dð U¸ð Uñ Uð Uñ „ðUð ñVð V˜tð V¸&ð Vñ Vð Vñ „ðVð ñSð S˜Dð S°vð Sñ Sð Sñ „ðSð ñRð R˜4ð R°fð Rñ Rð Rñ „ðRð ñSð S˜Dð S°vð Sñ Sð Sñ „ðSð ñTð T˜Tð T¸ð Tñ Tð Tñ „ðTð ñUð U˜dð U¸ð Uñ Uð Uñ „ðUð ñ]ð ] dð ]Àð ]ñ ]ð ]ñ „ð]ð ñ]ð ] dð ]ñ ]ð ]ñ „ð]ð ñ	Rð 	R˜4ð 	R°fð 	Rñ 	Rð 	Rñ „ð	Rð ñ	Sð 	S˜Dð 	S°vð 	Sñ 	Sð 	Sñ „ð	Sð ñ	Sð 	S˜Dð 	S°vð 	Sñ 	Sð 	Sñ „ð	Sð ñ	Sð 	S˜Dð 	S°vð 	Sñ 	Sð 	Sñ „ð	Sð ñ	Sð 	S˜Dð 	S°vð 	Sñ 	Sð 	Sñ „ð	Sð ñ	Sð 	S˜Dð 	S°vð 	Sñ 	Sð 	Sñ „ð	Sð ñ	Sð 	S˜Dð 	S°vð 	Sñ 	Sð 	Sñ „ð	Sð ñSð S˜Dð S°vð Sñ Sð Sñ „ðSð Ð ˜Hð ,'ÙÐ%ð,'áÐ%ð,'ñ Ð'ð,'ñ Ð'ð	,'ñ
 Ð4ð,'ñ ÐAð,'ñ ÐBð,'ñ ÐBð,'ñ ÐBð,'ñ Ð6ð,'ñ Ð6ð,'ñ ÐDð,'ñ ÐFð,'ñ ÐFð,'ñ ÐHð,'ñ  Ð4ð!,'ñ" Ð4ð#,'ð ,'ñ$ ÑBð%,'ñ& ÑDð','ñ( ÑDð),'ñ* ÑFð+,'ñ, Ñ/ð-,'ñ. Ñ/ð/,'ñ0 Ñ/ð1,'ñ2 Ñ1ð3,'ñ4 Ñ1ð5,'ñ6 Ñ0ð7,'ñ8 Ñ2ð9,'ñ: Ñ2ð;,'ñ< Ñ0ð=,'ñ> Ñ2ð?,'ñ@ Ñ2ðA,'ñB Ñ0ðC,'ñD Ñ2ðE,'ð ,'ñF 3Ù9Ù;Ù;Ù=Ù ?Ù ?Ù+Ù/ñW,'ð ,'ð ,'ñ ,ô ,ð ,ð ,ð ,r/   