QЦiR%SSKrSSKrSSKJr SSKJr SSKJrJrJ r J r J r J r J r Jr SSKrSSKJrJr SSKJr SSKJrJr SS KJrJr SS KJr S S KJrJrJr S S K J!r! S SK"J#r#J$r$J%r% /SQr&\"SS55r'"SS\'5r("SS\'5r)"SS\RT5r+"SS\RT5r,"SS\RT5r-S\ \\(\)4S\.S\ \/S\ \S \0S!\S"\-4S#jr1S$\2S!\S"\ \ \\(\)4\ \/44S%jr3S&\!0r4\ \2\4\5S''0\4ES(S)S*.Er60\4ES+S,S*.Er7"S-S.\5r8"S/S0\5r9"S1S2\5r:"S3S4\5r;"S5S6\5r<"S7S8\5r="S9S:\5r>"S;S<\5r?"S=S>\5r@"S?S@\5rA"SASB\5rB\"5\%"SC\8R4SD9SSESF.S\ \8S \0S!\S"\-4SGjj55rD\"5\%"SC\9R4SD9SSESF.S\ \9S \0S!\S"\-4SHjj55rE\"5\%"SC\:R4SD9SSESF.S\ \:S \0S!\S"\-4SIjj55rF\"5\%"SC\;R4SD9SSESF.S\ \;S \0S!\S"\-4SJjj55rG\"5\%"SC\R4SD9SSESF.S\ \>S \0S!\S"\-4SMjj55rJ\"5\%"SC\?R4SD9SSESF.S\ \?S \0S!\S"\-4SNjj55rK\"5\%"SC\@R4SD9SSESF.S\ \@S \0S!\S"\-4SOjj55rL\"5\%"SC\AR4SD9SSESF.S\ \AS \0S!\S"\-4SPjj55rM\"5\%"SC\BR4SD9SSESF.S\ \BS \0S!\S"\-4SQjj55rNg)RN) dataclass)partial)AnyCallableDictListOptionalSequenceTupleUnion)nnTensor)StochasticDepth)Conv2dNormActivationSqueezeExcitation)ImageClassificationInterpolationMode)_log_api_usage_once)register_modelWeights WeightsEnum)_IMAGENET_CATEGORIES)_make_divisible_ovewrite_named_paramhandle_legacy_interface) EfficientNetEfficientNet_B0_WeightsEfficientNet_B1_WeightsEfficientNet_B2_WeightsEfficientNet_B3_WeightsEfficientNet_B4_WeightsEfficientNet_B5_WeightsEfficientNet_B6_WeightsEfficientNet_B7_WeightsEfficientNet_V2_S_WeightsEfficientNet_V2_M_WeightsEfficientNet_V2_L_Weightsefficientnet_b0efficientnet_b1efficientnet_b2efficientnet_b3efficientnet_b4efficientnet_b5efficientnet_b6efficientnet_b7efficientnet_v2_sefficientnet_v2_mefficientnet_v2_lc \rSrSr%\\S'\\S'\\S'\\S'\\S'\\S'\S\R4\S '\ SS \S \S \ \S\4Sjj5r Sr g ) _MBConvConfig. expand_ratiokernelstrideinput_channels out_channels num_layers.blockNchannels width_mult min_valuereturnc [X-SU5$)N)r)r?r@rAs ^/var/www/html/ai-image-ml/venv/lib/python3.13/site-packages/torchvision/models/efficientnet.pyadjust_channels_MBConvConfig.adjust_channels8sx4aCCN)__name__ __module__ __qualname____firstlineno__float__annotations__intrr Module staticmethodr rF__static_attributes__rIrHrEr6r6.so K KO CN ##D#D5DXc]D^aDDrHr6c^\rSrSrSS\S\S\S\S\S\S \S \S \\S \R4S S4U4Sjjjr \ S\S \4Sj5r Sr U=r$) MBConvConfig=Nr8r9r:r;r<r=r@ depth_multr>.rBc >URXG5nURXW5nURXh5nU c[n [T U]XX4XVU 5 grJ)rF adjust_depthMBConvsuper__init__) selfr8r9r:r;r<r=r@rXr> __class__s rEr]MBConvConfig.__init__?sS--nI++LE &&z> =E v|afgrHcD[[R"X-55$rJ)rQmathceil)r=rXs rErZMBConvConfig.adjust_depthRs499Z4566rHrI)?reN)rKrLrMrNrOrQr rr rRr]rSrZrT __classcell__r_s@rErVrV=s 48hhh h  h  hhhhbii01h hh&77%77rHrVcv^\rSrSrSS\S\S\S\S\S\S \\S \R4S S4U4S jjjr S r U=r $)FusedMBConvConfigWNr8r9r:r;r<r=r>.rBc <>Uc[n[TU] XX4XVU5 grJ) FusedMBConvr\r]) r^r8r9r:r;r<r=r>r_s rEr]FusedMBConvConfig.__init__Ys# =E v|afgrHrIrJ) rKrLrMrNrOrQr rr rRr]rTrfrgs@rEririWs|59 h h h h  h  h hbii01 h  h hrHric ^\rSrSr\4S\S\S\S\R4S\S\R4SS4 U4S jjjr S \ S\ 4S jr S r U=r$) r[hcnfstochastic_depth_prob norm_layer.se_layerrBNc X>[T U]5 SURs=::aS::d O [S5eURS:H=(a URUR :HUl/n[RnURURUR5nXqR:wa&UR[URUSUUS95 UR[UUURURUUUS95 [SURS-5nURU"Xx[[RSS9S 95 UR[XqR SUSS95 [R "U6Ul[%US 5UlUR Ulg) Nrrillegal stride value kernel_sizerractivation_layer)rwr:groupsrrrxT)inplace) activationrow)r\r]r: ValueErrorr;r<use_res_connectr SiLUrFr8appendrr9maxr Sequentialr>rstochastic_depth) r^rprqrrrslayersrxexpanded_channelssqueeze_channelsr_s rEr]MBConv.__init__is~ SZZ$1$34 4"zzQY33E3EIYIY3Y"$77 //0B0BCDTDTU  2 2 2 MM$&&% !)%5    !!JJzz(%!1  q#"4"4"9: h0wWYW^W^hlOmno  !#3#3zlp  ]]F+ /0Eu M,,rHinputctURU5nUR(aURU5nX!- nU$rJr>rrr^rresults rEforwardMBConv.forward5E"   **62F OF rHr>r<rr)rKrLrMrNrrVrOrr rRr]rrrTrfrgs@rEr[r[hsu .? 8- 8- %8-S"))^, 8- 3 >* 8-  8-8-tVrHr[cl^\rSrSrS\S\S\S\R4SS4U4Sjjr S \ S\ 4S jr S r U=r $) rlrprqrr.rBNc  >[TU]5 SURs=::aS::d O [S5eURS:H=(a URUR :HUl/n[RnURURUR5nXaR:waaUR[URUURURUUS95 UR[XaR SUSS95 OEUR[URUR URURUUS95 [R"U6Ul[!US5UlUR Ulg)Nrrrurwr:rrrxrvr})r\r]r:r~r;r<rr rrFr8rrr9rr>rr)r^rprqrrrrxrr_s rEr]FusedMBConv.__init__sT SZZ$1$34 4"zzQY33E3EIYIY3Y"$77//0B0BCDTDTU  2 2 2 MM$&&% # ::)%5   MM$%'7'7QS]pt  MM$&&$$ # ::)%5  ]]F+ /0Eu M,,rHrctURU5nUR(aURU5nX!- nU$rJrrs rErFusedMBConv.forwardrrHr)rKrLrMrNrirOrr rRr]rrrTrfrgs@rErlrlsT2- 2- %2-S"))^, 2-  2-hVrHrlc^\rSrSrSS\\\\4S\S\S\ S\ \ S\ R4S \ \ S S4U4S jjjrS \S \4S jrS \S \4SjrSrU=r$)rNinverted_residual_settingdropoutrq num_classesrr. last_channelrBc >[TU]5 [U5 U(d [S5e[ U[ 5(a/[ UVs/sHn[ U[5PM sn5(d [S5eUc[Rn/nUSRn UR[SU SSU[RS95 [SU55n Sn UHn /n [!U R"5Hpn[$R$"U 5nU (aUR&Ul S UlU[+U 5-U - nU RUR-UUU55 U S - n Mr UR[R."U 65 M US R&nUbUOS U-nUR[UUS U[RS 95 [R."U6Ul[R2"S 5Ul[R."[R6"US S9[R8"UU55UlUR=5GHn[ U[R>5(ab[R@RCURDSS9 URFb+[R@RIURF5 MM[ U[R[RJ45(aU[R@RMURD5 [R@RIURF5 GM [ U[R85(dGM+S[NRP"URR5- n[R@RUURDU*U5 [R@RIURF5 GM gs snf)a EfficientNet V1 and V2 main class Args: inverted_residual_setting (Sequence[Union[MBConvConfig, FusedMBConvConfig]]): Network structure dropout (float): The droupout probability stochastic_depth_prob (float): The stochastic depth probability num_classes (int): Number of classes norm_layer (Optional[Callable[..., nn.Module]]): Module specifying the normalization layer to use last_channel (int): The number of channels on the penultimate layer z1The inverted_residual_setting should not be emptyz:The inverted_residual_setting should be List[MBConvConfig]Nrrrc38# UHoRv M g7frJ)r=).0rps rE (EfficientNet.__init__..s U;TC;TsrrzrvT)pr{fan_out)modere)+r\r]rr~ isinstancer allr6 TypeErrorr BatchNorm2dr;rrrsumranger=copyr<r:rOr>rfeaturesAdaptiveAvgPool2davgpoolDropoutLinear classifiermodulesConv2dinitkaiming_normal_weightbiaszeros_ GroupNormones_rbsqrt out_featuresuniform_)r^rrrqrrrrsrfirstconv_output_channelstotal_stage_blocksstage_block_idrpstage_ block_cnfsd_problastconv_input_channelslastconv_output_channelsm init_ranger_s rEr]EfficientNet.__init__s ( D!(PQ Q 0( ; ;;TU;TaZ=1;TUVVXY Y  J"$%>a$@$O$O! ,!AR\oqovov  ! U;T UU,C%'E3>>* IIcN /8/E/EI,'(I$0%2GGJ\\ Y__YLM!#+ MM"--/ 0#-(#> ahh'qvv&Aryy)) 499Q^^#<<   J; Cqvv& wVs PxcURU5nURU5n[R"US5nUR U5nU$)Nr)rrtorchflattenrr^rs rE _forward_implEfficientNet._forward_implLs@ MM!  LLO MM!Q  OOA rHc$URU5$rJ)rrs rErEfficientNet.forwardVs!!!$$rH)rrr)皙?iNN)rKrLrMrNr r rVrirOrQr rr rRr]rrrrTrfrgs@rErrs (+9=&*a'#+E,@Q2Q,R#Sa'a' % a'  a' Xc299n56 a'sma' a'a'Fv&%%F%%rHrrrrweightsprogresskwargsrBc Ub#[US[URS55 [X4SU0UD6nUbUR UR USS95 U$)Nr categoriesrT)r check_hash)rlenmetarload_state_dictget_state_dict)rrrrrrmodels rE _efficientnetrZsdfmSl9S5TU 2 a, aZ` aE g44hSW4XY LrHarchc `URS5(a[[URS5URS5S9nU"SSSSSS5U"S SS SS S 5U"S S S S S S 5U"S SS S SS5U"S S SSSS5U"S S S SSS5U"S SSSSS5/nSnX44$URS5(aa[ SSSS S S 5[ SSS S SS5[ SSS SSS5[SSS SSS 5[S SSSSS5[S SS SSS5/nSnX44$URS5(ap[ SSSS S S5[ SSS S SS 5[ SSS SSS 5[SSS SSS5[S SSSSS5[S SS SS S!5[S SSS S"S 5/nSnX44$URS#5(ap[ SSSSSS5[ SSS SSS5[ SSS SS$S5[SSS S$SS%5[S SSSS&S'5[S SS S&S(S)5[S SSS(S*S5/nSnX44$[ S+U35e),Nefficientnet_br@rXr@rXrr r(Pprz@r20@ ir3i0ir4` izUnsupported model type ) startswithrrVpoprir~)rr bneck_confrrs rE_efficientnet_confrms  '((\fjj6N[a[e[efr[st q!QB * q!QB * q!QB * q!QB * q!QC + q!QS! , q!QS! ,% ! H % 22G , - - aAr2q 1 aAr2q 1 aAr2q 1 Aq"c1 - Aq#sA . Aq#sB / % ! 4 % 223 , - - aAr2q 1 aAr2q 1 aAr2q 1 Aq"c1 - Aq#sB / Aq#sB / Aq#sA .% !  % 22 , - - aAr2q 1 aAr2q 1 aAr2q 1 Aq"c2 . Aq#sB / Aq#sB / Aq#sA .% !  % 2224&9::rHr _COMMON_META)rrzUhttps://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v1)min_sizerecipe)!r zUhttps://github.com/pytorch/vision/tree/main/references/classification#efficientnet-v2c h\rSrSr\"S\"\SS\RS90\ ESSSS S .0S S S S.ES9r \ r Sr g)rizJhttps://download.pytorch.org/models/efficientnet_b0_rwightman-7f5810bc.pthrr crop_size resize_size interpolationidP ImageNet-1Kg?5^IlS@g5^I bW@zacc@1zacc@5gNbX9?g~jts4@1These weights are ported from the original paper. num_params_metrics_ops _file_size_docsurl transformsrrIN rKrLrMrNrrrrBICUBIC_COMMON_META_V1 IMAGENET1K_V1DEFAULTrTrIrHrErra X 3CO`OhOh     !##   L  M(GrHrc \rSrSr\"S\"\SS\RS90\ ESSSS S .0S S S S.ES9r \"S\"\SS\RS90\ ESSSSSS .0S SSS.ES9r \ r Srg)r izJhttps://download.pytorch.org/models/efficientnet_b1_rwightman-bac287d4.pthrr ivrg +S@gClW@rgCl?gM">@rrrz@https://download.pytorch.org/models/efficientnet_b1-c27df63c.pthzOhttps://github.com/pytorch/vision/issues/3995#new-recipe-with-lr-wd-crop-tuninggʡS@gƻW@gA`">@$ These weights improve upon the results of the original paper by using a modified version of TorchVision's `new training recipe `_. )rr rrrrrIN)rKrLrMrNrrrrrrr BILINEAR IMAGENET1K_V2r!rTrIrHrEr r s X 3CO`OhOh     !##   L  M( N 3CO`OiOi   !g##    M0GrHr c h\rSrSr\"S\"\SS\RS90\ ESSSSS .0S S S S .ES9r \ r Sr g)r!izJhttps://download.pytorch.org/models/efficientnet_b2_rwightman-c35c1473.pthi r irgx&T@gp= W@rg rh?gʡEA@rrrrINrrIrHrEr!r!r"rHr!c h\rSrSr\"S\"\SS\RS90\ ESSSS S .0S S S S.ES9r \ r Sr g)r"izJhttps://download.pytorch.org/models/efficientnet_b3_rwightman-b3899882.pthi,rr irgnT@g~jtX@rgZd;?gd;OG@rrrrINrrIrHrEr"r"a X 3CO`OhOh     "##   L  M(GrHr"c h\rSrSr\"S\"\SS\RS90\ ESSSS S .0S S S S.ES9r \ r Sr g)r#i/zJhttps://download.pytorch.org/models/efficientnet_b4_rwightman-23ab8bcd.pthi|rr i0!'rgjtT@gt&X@rg~jt@gKR@rrrrINrrIrHrEr#r#/r+rHr#c h\rSrSr\"S\"\SS\RS90\ ESSSSS .0S S S S .ES9r \ r Sr g)r$iGzJhttps://download.pytorch.org/models/efficientnet_b5_lukemelas-1a07897c.pthir irg#~jT@gx&1(X@rgx&1$@gK7]@rrrrINrrIrHrEr$r$Ga X 3CO`OhOh     "##  !L  M(GrHr$c h\rSrSr\"S\"\SS\RS90\ ESSSSS .0S S S S .ES9r \ r Sr g)r%i_zJhttps://download.pytorch.org/models/efficientnet_b6_lukemelas-24a108a5.pthir irgnU@gv:X@rg rh3@g$d@rrrrINrrIrHrEr%r%_r.rHr%c h\rSrSr\"S\"\SS\RS90\ ESSSSS .0S S S S .ES9r \ r Sr g)r&iwzJhttps://download.pytorch.org/models/efficientnet_b7_lukemelas-c5b4e57e.pthiXr icrg+U@g'1:X@rgsh|B@go@rrrrINrrIrHrEr&r&wr.rHr&c h\rSrSr\"S\"\SS\RS90\ ESSSSS .0S S S S .ES9r \ r Sr g)r'izBhttps://download.pytorch.org/models/efficientnet_v2_s-dd5fe13b.pthrr i8nGrg;OU@gx&18X@rgZd @gVT@r&rrrIN rKrLrMrNrrrrr'_COMMON_META_V2r r!rTrIrHrEr'r'se P +44    "##    M4GrHr'c h\rSrSr\"S\"\SS\RS90\ ESSSSS .0S S S S .ES9r \ r Sr g)r(izBhttps://download.pytorch.org/models/efficientnet_v2_m-dc08266a.pthr i:rgI +GU@gDlIX@rgE8@gQj@r&rrrINr2rIrHrEr(r(se P +44    "##    M4GrHr(c l\rSrSr\"S\"\SS\RSSS90\ ESSSS S .0S S S S.ES9r \ r Sr g)r)izBhttps://download.pytorch.org/models/efficientnet_v2_l-59c71312.pthr5)?r7r7)rrrmeanstdiHfrgʡEsU@gOnrX@rg ףp= L@gI +i|@rrrrIN) rKrLrMrNrrrrrr3r r!rTrIrHrEr)r)si P +33       ###  !L  M0GrHr) pretrained)rT)rrc [RU5n[SSSS9up4[X2R SS5X@U40UD6$)aEfficientNet B0 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks `_ paper. Args: weights (:class:`~torchvision.models.EfficientNet_B0_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_B0_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_B0_Weights :members: r*rerrr)rverifyrrrrrrrrs rEr*r*Q.&,,W5G.@AR_bor.s+ !::i#=|V^ bh rHc [RU5n[SSSS9up4[X2R SS5X@U40UD6$)aEfficientNet B1 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks `_ paper. Args: weights (:class:`~torchvision.models.EfficientNet_B1_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_B1_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_B1_Weights :members: r+re皙?rrr)r r<rrrr=s rEr+r+r>rHc [RU5n[SSSS9up4[X2R SS5X@U40UD6$)aEfficientNet B2 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks `_ paper. Args: weights (:class:`~torchvision.models.EfficientNet_B2_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_B2_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_B2_Weights :members: r,r@333333?rr333333?)r!r<rrrr=s rEr,r,&r>rHc [RU5n[SSSS9up4[UUR SS5UUU40UD6$)aEfficientNet B3 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks `_ paper. Args: weights (:class:`~torchvision.models.EfficientNet_B3_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_B3_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_B3_Weights :members: r-rBffffff?rrrC)r"r<rrrr=s rEr-r-EZ.&,,W5G.@AR_bor.s+ ! 9c"     rHc [RU5n[SSSS9up4[UUR SS5UUU40UD6$)aEfficientNet B4 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks `_ paper. Args: weights (:class:`~torchvision.models.EfficientNet_B4_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_B4_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_B4_Weights :members: r.rE?rr皙?)r#r<rrrr=s rEr.r.irFrHc [RU5n[SSSS9up4[UUR SS5UUU4S[ [ RSS S 90UD6$) aEfficientNet B5 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks `_ paper. Args: weights (:class:`~torchvision.models.EfficientNet_B5_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_B5_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_B5_Weights :members: r/g?g@rrrIrrMbP?{Gz?epsmomentum)r$r<rrrrr rr=s rEr/r/q.&,,W5G.@AR_bor.s+ ! 9c"   2>>utD    rHc [RU5n[SSSS9up4[UUR SS5UUU4S[ [ RSS S 90UD6$) aEfficientNet B6 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks `_ paper. Args: weights (:class:`~torchvision.models.EfficientNet_B6_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_B6_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_B6_Weights :members: r0rHg@rrr7rrrKrLrM)r%r<rrrrr rr=s rEr0r0rPrHc [RU5n[SSSS9up4[UUR SS5UUU4S[ [ RSS S 90UD6$) aEfficientNet B7 model architecture from the `EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks `_ paper. Args: weights (:class:`~torchvision.models.EfficientNet_B7_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_B7_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_B7_Weights :members: r1g@g@rrr7rrrKrLrM)r&r<rrrrr rr=s rEr1r1rPrHc [RU5n[S5up4[UUR SS5UUU4S[ [ RSS90UD6$)a Constructs an EfficientNetV2-S architecture from `EfficientNetV2: Smaller Models and Faster Training `_. Args: weights (:class:`~torchvision.models.EfficientNet_V2_S_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_V2_S_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_V2_S_Weights :members: r2rrrrrKrN)r'r<rrrrr rr=s rEr2r2i0(..w7G.@AT.U+ ! 9c"   2>>u5    rHc [RU5n[S5up4[UUR SS5UUU4S[ [ RSS90UD6$)a Constructs an EfficientNetV2-M architecture from `EfficientNetV2: Smaller Models and Faster Training `_. Args: weights (:class:`~torchvision.models.EfficientNet_V2_M_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_V2_M_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_V2_M_Weights :members: r3rrCrrrKrT)r(r<rrrrr rr=s rEr3r3"rUrHc [RU5n[S5up4[UUR SS5UUU4S[ [ RSS90UD6$)a Constructs an EfficientNetV2-L architecture from `EfficientNetV2: Smaller Models and Faster Training `_. Args: weights (:class:`~torchvision.models.EfficientNet_V2_L_Weights`, optional): The pretrained weights to use. See :class:`~torchvision.models.EfficientNet_V2_L_Weights` below for more details, and possible values. By default, no pre-trained weights are used. progress (bool, optional): If True, displays a progress bar of the download to stderr. Default is True. **kwargs: parameters passed to the ``torchvision.models.efficientnet.EfficientNet`` base class. Please refer to the `source code `_ for more details about this class. .. autoclass:: torchvision.models.EfficientNet_V2_L_Weights :members: r4rrIrrrKrT)r)r<rrrrr rr=s rEr4r4HrUrH)Orrb dataclassesr functoolsrtypingrrrrr r r r rr rtorchvision.opsrops.miscrrtransforms._presetsrrutilsr_apirrr_metar_utilsrrr__all__r6rVrirRr[rlrrOrQboolrstrrrrPrr3rr r!r"r#r$r%r&r'r(r)r r*r+r,r-r.r/r0r1r2r3r4rIrHrEres !NNN +>H'66'SS 6  D D  D7=74h h"@RYY@F:")):zo%299o%d'l